Menu
▾
▴
hiccupmail-devel — List for anyone interested in developing the Hiccpmail project.
You can subscribe to this list here.
| 2001 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
(14) |
Aug
(14) |
Sep
|
Oct
|
Nov
|
Dec
|
|---|
|
From: David M. B. <da...@me...> - 2001-08-15 13:20:23
|
Hmm. Maybe you should ask our Yotam?
Clueless,
David
"Karl O . Pinc" <ko...@me...> said:
> Ok, if you can help, here's the message I sent to ai...@fr....
>
> The latest update is the scsi subsystem seens to hang after a while doing
> this on 2.4.8.
>
> (Well, you asked about helping. :-)
>
> On 2001.08.14 17:07 Karl O . Pinc wrote:
> Hi, I'm a newbie. Spank me. :-) Please let me know what further info is
> needed, etc.
>
> I've got a scsi board labeled AHA2840A/42A that AFIK worked fine under
> linux 2.0.34 for years. My first question is this:
>
> Is the AHA284xA supported?
>
> The README.aic7xxx says that the AHA284x is supported, but makes no mention
> of AHA284xA. AFIK, the difference is that the 284x has jumpers, the 284xA
> has dip switches. The 284xA dip switches control: I/O port address, floppy
> enable/disable, BIOS address, and BIOS enable/disable. I recall the
> jumpers on the 284x can at least also en/disable scsi target disconnect,
> sync negoiation and probably some other stuff. I _think_ all this is now
> set in the card's bios cmos.
>
> Assuming the card's supported, here's what's happening:
>
> I'm re-incarnating the box and decided to load test it (copying large files
> from drive to drive. 3 drives, 4 processes, file sizes of {2,3,5,7} *
> 10^7B, 64MB ram). Running linux-2.4.7, I get a rare (every few hours)
> "aic7xxx_abort return 8194", I've seen it while attempting to send ABORT to
> a drive. (Sorry, I'm running 2.4.7 on top of the old 2.0.34 system, just
> to test, and syslogd seems to be down, so I don't have a written log to
> work from here. Better info collected below.)
>
> I just switched to 2.4.8 and I get frequent (most of the time, there's no
> disk activity while the driver issues errors) "aic7xxx_abort return 8194"
> messages. Mostly on ABORT but sometimes on TARGET_RESET. There are so
> many the driver goes on to do "Recovery SCB completes" type messages.
>
> Where do I go from here?
>
> (FYI, Starting about 12 hours from now, I won't have access to the hardware
> for a few weeks.)
>
> Here are details from /proc:
>
> ----after boot of 2.4.7----------
> [root@arthur /root]# uname -a
> Linux arthur.meme.com 2.4.7 #1 Mon Aug 6 11:30:42 CDT 2001 i486 unknown
> [root@arthur /root]# cat /proc/scsi/aic7xxx/0
> Adaptec AIC7xxx driver version: 6.1.13
> aic7770: Single Channel A, SCSI Id=7, 4/255 SCBs
> Channel A Target 0 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Goal: 10.000MB/s transfers (10.000MHz, offset 15)
> Curr: 10.000MB/s transfers (10.000MHz, offset 15)
> Channel A Target 0 Lun 0 Settings
> Commands Queued 4
> Commands Active 0
> Command Openings 253
> Max Tagged Openings 253
> Device Queue Frozen Count 0
> Channel A Target 1 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 2 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 3 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 4 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Goal: 3.300MB/s transfers
> Curr: 3.300MB/s transfers
> Channel A Target 4 Lun 0 Settings
> Commands Queued 1383
> Commands Active 1
> Command Openings 0
> Max Tagged Openings 0
> Device Queue Frozen Count 0
> Channel A Target 5 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Goal: 10.000MB/s transfers (10.000MHz, offset 15)
> Curr: 10.000MB/s transfers (10.000MHz, offset 15)
> Channel A Target 5 Lun 0 Settings
> Commands Queued 5
> Commands Active 0
> Command Openings 253
> Max Tagged Openings 253
> Device Queue Frozen Count 0
> Channel A Target 6 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 7 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> [root@arthur /root]# cat /proc/scsi/scsi
> Attached devices:
> Host: scsi0 Channel: 00 Id: 00 Lun: 00
> Vendor: QUANTUM Model: FIREBALL ST4300S Rev: 0F0D
> Type: Direct-Access ANSI SCSI revision: 02
> Host: scsi0 Channel: 00 Id: 04 Lun: 00
> Vendor: IMPRIMIS Model: 94601-15 Rev: 4202
> Type: Direct-Access ANSI SCSI revision: 01 CCS
> Host: scsi0 Channel: 00 Id: 05 Lun: 00
> Vendor: CONNER Model: CFP1060S 1.05GB Rev: 245F
> Type: Direct-Access ANSI SCSI revision: 02
>
>
> ---------Begin lots-o-copy, console says:
> (scsi:0:A:0:0): Locking max tag count at 8
>
> After a half hour or so of copying, there's still no error and this is what
> proc looks like:
>
> [root@arthur /root]# cat /proc/scsi/aic7xxx/0
> Adaptec AIC7xxx driver version: 6.1.13
> aic7770: Single Channel A, SCSI Id=7, 4/255 SCBs
> Channel A Target 0 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Goal: 10.000MB/s transfers (10.000MHz, offset 15)
> Curr: 10.000MB/s transfers (10.000MHz, offset 15)
> Channel A Target 0 Lun 0 Settings
> Commands Queued 56765
> Commands Active 0
> Command Openings 8
> Max Tagged Openings 8
> Device Queue Frozen Count 0
> Channel A Target 1 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 2 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 3 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 4 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Goal: 3.300MB/s transfers
> Curr: 3.300MB/s transfers
> Channel A Target 4 Lun 0 Settings
> Commands Queued 34178
> Commands Active 1
> Command Openings 0
> Max Tagged Openings 0
> Device Queue Frozen Count 0
> Channel A Target 5 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Goal: 10.000MB/s transfers (10.000MHz, offset 15)
> Curr: 10.000MB/s transfers (10.000MHz, offset 15)
> Channel A Target 5 Lun 0 Settings
> Commands Queued 36040
> Commands Active 2
> Command Openings 31
> Max Tagged Openings 253
> Device Queue Frozen Count 0
> Channel A Target 6 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 7 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
>
> ---------------Rebooting with 2.4.8, /proc is:
>
>
> [root@arthur /root]# uname -a
> Linux arthur.meme.com 2.4.8 #3 Tue Aug 14 14:06:17 CDT 2001 i486 unknown
> [root@arthur /root]# cat /proc/scsi/aic7xxx/0
> Adaptec AIC7xxx driver version: 6.1.13
> aic7770: Single Channel A, SCSI Id=7, 4/255 SCBs
> Channel A Target 0 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Goal: 10.000MB/s transfers (10.000MHz, offset 15)
> Curr: 10.000MB/s transfers (10.000MHz, offset 15)
> Channel A Target 0 Lun 0 Settings
> Commands Queued 25
> Commands Active 0
> Command Openings 253
> Max Tagged Openings 253
> Device Queue Frozen Count 0
> Channel A Target 1 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 2 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 3 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 4 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Goal: 3.300MB/s transfers
> Curr: 3.300MB/s transfers
> Channel A Target 4 Lun 0 Settings
> Commands Queued 1496
> Commands Active 0
> Command Openings 1
> Max Tagged Openings 0
> Device Queue Frozen Count 0
> Channel A Target 5 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Goal: 10.000MB/s transfers (10.000MHz, offset 15)
> Curr: 10.000MB/s transfers (10.000MHz, offset 15)
> Channel A Target 5 Lun 0 Settings
> Commands Queued 11
> Commands Active 0
> Command Openings 253
> Max Tagged Openings 253
> Device Queue Frozen Count 0
> Channel A Target 6 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 7 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
>
> [root@arthur /root]# cat /proc/scsi/scsi
> Attached devices:
> Host: scsi0 Channel: 00 Id: 00 Lun: 00
> Vendor: QUANTUM Model: FIREBALL ST4300S Rev: 0F0D
> Type: Direct-Access ANSI SCSI revision: 02
> Host: scsi0 Channel: 00 Id: 04 Lun: 00
> Vendor: IMPRIMIS Model: 94601-15 Rev: 4202
> Type: Direct-Access ANSI SCSI revision: 01 CCS
> Host: scsi0 Channel: 00 Id: 05 Lun: 00
> Vendor: CONNER Model: CFP1060S 1.05GB Rev: 245F
> Type: Direct-Access ANSI SCSI revision: 02
>
>
>
> scsi0:0:0:0: Attempting to queue an ABORT message
> scsi0:0:0:0: Command found on device queue
> aic7xxx_abort_returns 8194
>
> -(FWIW, I recall seeing that it was scsci id 5, not 0 that was causing the
> messages under linux 2.4.7.)
>
> -repeat about about 10 times then
>
> scsi0:0:0:0: Attempting to queue a TARGET RESET message
>
> -is the message instead of the "ABORT" line
>
> -No disk activity occurs while all this is going on.
>
> -Then, there's mixed in message about SCB recovery (Sorry, I'm copying off
> the console as messages scroll by so they are not in the correct order.)
>
> scsi0:A:0:0 Queuing a recovery SCB
> Recovery code sleeping
> Recovery SCB completes
> Recovery code awake
>
> scsi0:0:0: Abort Tag Message Sent
> SCB7 - Abort Tag Completed.
>
> ------------- Here's what /proc says while all this is going on (still
> 2.4.8):
>
>
> [root@arthur /root]# cat /proc/scsi/aic7xxx/0
> Adaptec AIC7xxx driver version: 6.1.13
> aic7770: Single Channel A, SCSI Id=7, 4/255 SCBs
> Channel A Target 0 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Goal: 10.000MB/s transfers (10.000MHz, offset 15)
> Curr: 10.000MB/s transfers (10.000MHz, offset 15)
> Channel A Target 0 Lun 0 Settings
> Commands Queued 5258
> Commands Active 8
> Command Openings 0
> Max Tagged Openings 253
> Device Queue Frozen Count 0
> Channel A Target 1 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 2 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 3 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 4 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Goal: 3.300MB/s transfers
> Curr: 3.300MB/s transfers
> Channel A Target 4 Lun 0 Settings
> Commands Queued 4823
> Commands Active 0
> Command Openings 1
> Max Tagged Openings 0
> Device Queue Frozen Count 0
> Channel A Target 5 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Goal: 10.000MB/s transfers (10.000MHz, offset 15)
> Curr: 10.000MB/s transfers (10.000MHz, offset 15)
> Channel A Target 5 Lun 0 Settings
> Commands Queued 110
> Commands Active 0
> Command Openings 253
> Max Tagged Openings 253
> Device Queue Frozen Count 0
> Channel A Target 6 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 7 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
>
> ----and again, just to get another snapshot:
>
>
> [root@arthur /root]# cat /proc/scsi/aic7xxx/0
> Adaptec AIC7xxx driver version: 6.1.13
> aic7770: Single Channel A, SCSI Id=7, 4/255 SCBs
> Channel A Target 0 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Goal: 10.000MB/s transfers (10.000MHz, offset 15)
> Curr: 10.000MB/s transfers (10.000MHz, offset 15)
> Channel A Target 0 Lun 0 Settings
> Commands Queued 7608
> Commands Active 8
> Command Openings 0
> Max Tagged Openings 253
> Device Queue Frozen Count 0
> Channel A Target 1 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 2 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 3 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 4 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Goal: 3.300MB/s transfers
> Curr: 3.300MB/s transfers
> Channel A Target 4 Lun 0 Settings
> Commands Queued 7193
> Commands Active 1
> Command Openings 0
> Max Tagged Openings 0
> Device Queue Frozen Count 0
> Channel A Target 5 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Goal: 10.000MB/s transfers (10.000MHz, offset 15)
> Curr: 10.000MB/s transfers (10.000MHz, offset 15)
> Channel A Target 5 Lun 0 Settings
> Commands Queued 854
> Commands Active 0
> Command Openings 253
> Max Tagged Openings 253
> Device Queue Frozen Count 0
> Channel A Target 6 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
> Channel A Target 7 Negotiation Settings
> User: 10.000MB/s transfers (10.000MHz, offset 255)
>
>
> Thanks,
>
> Karl <ko...@me...>
>
> _______________________________________________
> hiccupmail-devel mailing list
> hic...@li...
> http://lists.sourceforge.net/lists/listinfo/hiccupmail-devel
>
--
David M. Bucknell
http://members.iteachnet.org/~david/oldweb
http://www.OpenSourceSchools.org
International Education Daily
http://members.iteachnet.org/webzine/
|
|
From: Karl O . P. <ko...@me...> - 2001-08-14 22:35:11
|
Ok, if you can help, here's the message I sent to ai...@fr....
The latest update is the scsi subsystem seens to hang after a while doing
this on 2.4.8.
(Well, you asked about helping. :-)
On 2001.08.14 17:07 Karl O . Pinc wrote:
Hi, I'm a newbie. Spank me. :-) Please let me know what further info is
needed, etc.
I've got a scsi board labeled AHA2840A/42A that AFIK worked fine under
linux 2.0.34 for years. My first question is this:
Is the AHA284xA supported?
The README.aic7xxx says that the AHA284x is supported, but makes no mention
of AHA284xA. AFIK, the difference is that the 284x has jumpers, the 284xA
has dip switches. The 284xA dip switches control: I/O port address, floppy
enable/disable, BIOS address, and BIOS enable/disable. I recall the
jumpers on the 284x can at least also en/disable scsi target disconnect,
sync negoiation and probably some other stuff. I _think_ all this is now
set in the card's bios cmos.
Assuming the card's supported, here's what's happening:
I'm re-incarnating the box and decided to load test it (copying large files
from drive to drive. 3 drives, 4 processes, file sizes of {2,3,5,7} *
10^7B, 64MB ram). Running linux-2.4.7, I get a rare (every few hours)
"aic7xxx_abort return 8194", I've seen it while attempting to send ABORT to
a drive. (Sorry, I'm running 2.4.7 on top of the old 2.0.34 system, just
to test, and syslogd seems to be down, so I don't have a written log to
work from here. Better info collected below.)
I just switched to 2.4.8 and I get frequent (most of the time, there's no
disk activity while the driver issues errors) "aic7xxx_abort return 8194"
messages. Mostly on ABORT but sometimes on TARGET_RESET. There are so
many the driver goes on to do "Recovery SCB completes" type messages.
Where do I go from here?
(FYI, Starting about 12 hours from now, I won't have access to the hardware
for a few weeks.)
Here are details from /proc:
----after boot of 2.4.7----------
[root@arthur /root]# uname -a
Linux arthur.meme.com 2.4.7 #1 Mon Aug 6 11:30:42 CDT 2001 i486 unknown
[root@arthur /root]# cat /proc/scsi/aic7xxx/0
Adaptec AIC7xxx driver version: 6.1.13
aic7770: Single Channel A, SCSI Id=7, 4/255 SCBs
Channel A Target 0 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Goal: 10.000MB/s transfers (10.000MHz, offset 15)
Curr: 10.000MB/s transfers (10.000MHz, offset 15)
Channel A Target 0 Lun 0 Settings
Commands Queued 4
Commands Active 0
Command Openings 253
Max Tagged Openings 253
Device Queue Frozen Count 0
Channel A Target 1 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 2 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 3 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 4 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Goal: 3.300MB/s transfers
Curr: 3.300MB/s transfers
Channel A Target 4 Lun 0 Settings
Commands Queued 1383
Commands Active 1
Command Openings 0
Max Tagged Openings 0
Device Queue Frozen Count 0
Channel A Target 5 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Goal: 10.000MB/s transfers (10.000MHz, offset 15)
Curr: 10.000MB/s transfers (10.000MHz, offset 15)
Channel A Target 5 Lun 0 Settings
Commands Queued 5
Commands Active 0
Command Openings 253
Max Tagged Openings 253
Device Queue Frozen Count 0
Channel A Target 6 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 7 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
[root@arthur /root]# cat /proc/scsi/scsi
Attached devices:
Host: scsi0 Channel: 00 Id: 00 Lun: 00
Vendor: QUANTUM Model: FIREBALL ST4300S Rev: 0F0D
Type: Direct-Access ANSI SCSI revision: 02
Host: scsi0 Channel: 00 Id: 04 Lun: 00
Vendor: IMPRIMIS Model: 94601-15 Rev: 4202
Type: Direct-Access ANSI SCSI revision: 01 CCS
Host: scsi0 Channel: 00 Id: 05 Lun: 00
Vendor: CONNER Model: CFP1060S 1.05GB Rev: 245F
Type: Direct-Access ANSI SCSI revision: 02
---------Begin lots-o-copy, console says:
(scsi:0:A:0:0): Locking max tag count at 8
After a half hour or so of copying, there's still no error and this is what
proc looks like:
[root@arthur /root]# cat /proc/scsi/aic7xxx/0
Adaptec AIC7xxx driver version: 6.1.13
aic7770: Single Channel A, SCSI Id=7, 4/255 SCBs
Channel A Target 0 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Goal: 10.000MB/s transfers (10.000MHz, offset 15)
Curr: 10.000MB/s transfers (10.000MHz, offset 15)
Channel A Target 0 Lun 0 Settings
Commands Queued 56765
Commands Active 0
Command Openings 8
Max Tagged Openings 8
Device Queue Frozen Count 0
Channel A Target 1 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 2 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 3 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 4 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Goal: 3.300MB/s transfers
Curr: 3.300MB/s transfers
Channel A Target 4 Lun 0 Settings
Commands Queued 34178
Commands Active 1
Command Openings 0
Max Tagged Openings 0
Device Queue Frozen Count 0
Channel A Target 5 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Goal: 10.000MB/s transfers (10.000MHz, offset 15)
Curr: 10.000MB/s transfers (10.000MHz, offset 15)
Channel A Target 5 Lun 0 Settings
Commands Queued 36040
Commands Active 2
Command Openings 31
Max Tagged Openings 253
Device Queue Frozen Count 0
Channel A Target 6 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 7 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
---------------Rebooting with 2.4.8, /proc is:
[root@arthur /root]# uname -a
Linux arthur.meme.com 2.4.8 #3 Tue Aug 14 14:06:17 CDT 2001 i486 unknown
[root@arthur /root]# cat /proc/scsi/aic7xxx/0
Adaptec AIC7xxx driver version: 6.1.13
aic7770: Single Channel A, SCSI Id=7, 4/255 SCBs
Channel A Target 0 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Goal: 10.000MB/s transfers (10.000MHz, offset 15)
Curr: 10.000MB/s transfers (10.000MHz, offset 15)
Channel A Target 0 Lun 0 Settings
Commands Queued 25
Commands Active 0
Command Openings 253
Max Tagged Openings 253
Device Queue Frozen Count 0
Channel A Target 1 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 2 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 3 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 4 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Goal: 3.300MB/s transfers
Curr: 3.300MB/s transfers
Channel A Target 4 Lun 0 Settings
Commands Queued 1496
Commands Active 0
Command Openings 1
Max Tagged Openings 0
Device Queue Frozen Count 0
Channel A Target 5 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Goal: 10.000MB/s transfers (10.000MHz, offset 15)
Curr: 10.000MB/s transfers (10.000MHz, offset 15)
Channel A Target 5 Lun 0 Settings
Commands Queued 11
Commands Active 0
Command Openings 253
Max Tagged Openings 253
Device Queue Frozen Count 0
Channel A Target 6 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 7 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
[root@arthur /root]# cat /proc/scsi/scsi
Attached devices:
Host: scsi0 Channel: 00 Id: 00 Lun: 00
Vendor: QUANTUM Model: FIREBALL ST4300S Rev: 0F0D
Type: Direct-Access ANSI SCSI revision: 02
Host: scsi0 Channel: 00 Id: 04 Lun: 00
Vendor: IMPRIMIS Model: 94601-15 Rev: 4202
Type: Direct-Access ANSI SCSI revision: 01 CCS
Host: scsi0 Channel: 00 Id: 05 Lun: 00
Vendor: CONNER Model: CFP1060S 1.05GB Rev: 245F
Type: Direct-Access ANSI SCSI revision: 02
scsi0:0:0:0: Attempting to queue an ABORT message
scsi0:0:0:0: Command found on device queue
aic7xxx_abort_returns 8194
-(FWIW, I recall seeing that it was scsci id 5, not 0 that was causing the
messages under linux 2.4.7.)
-repeat about about 10 times then
scsi0:0:0:0: Attempting to queue a TARGET RESET message
-is the message instead of the "ABORT" line
-No disk activity occurs while all this is going on.
-Then, there's mixed in message about SCB recovery (Sorry, I'm copying off
the console as messages scroll by so they are not in the correct order.)
scsi0:A:0:0 Queuing a recovery SCB
Recovery code sleeping
Recovery SCB completes
Recovery code awake
scsi0:0:0: Abort Tag Message Sent
SCB7 - Abort Tag Completed.
------------- Here's what /proc says while all this is going on (still
2.4.8):
[root@arthur /root]# cat /proc/scsi/aic7xxx/0
Adaptec AIC7xxx driver version: 6.1.13
aic7770: Single Channel A, SCSI Id=7, 4/255 SCBs
Channel A Target 0 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Goal: 10.000MB/s transfers (10.000MHz, offset 15)
Curr: 10.000MB/s transfers (10.000MHz, offset 15)
Channel A Target 0 Lun 0 Settings
Commands Queued 5258
Commands Active 8
Command Openings 0
Max Tagged Openings 253
Device Queue Frozen Count 0
Channel A Target 1 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 2 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 3 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 4 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Goal: 3.300MB/s transfers
Curr: 3.300MB/s transfers
Channel A Target 4 Lun 0 Settings
Commands Queued 4823
Commands Active 0
Command Openings 1
Max Tagged Openings 0
Device Queue Frozen Count 0
Channel A Target 5 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Goal: 10.000MB/s transfers (10.000MHz, offset 15)
Curr: 10.000MB/s transfers (10.000MHz, offset 15)
Channel A Target 5 Lun 0 Settings
Commands Queued 110
Commands Active 0
Command Openings 253
Max Tagged Openings 253
Device Queue Frozen Count 0
Channel A Target 6 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 7 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
----and again, just to get another snapshot:
[root@arthur /root]# cat /proc/scsi/aic7xxx/0
Adaptec AIC7xxx driver version: 6.1.13
aic7770: Single Channel A, SCSI Id=7, 4/255 SCBs
Channel A Target 0 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Goal: 10.000MB/s transfers (10.000MHz, offset 15)
Curr: 10.000MB/s transfers (10.000MHz, offset 15)
Channel A Target 0 Lun 0 Settings
Commands Queued 7608
Commands Active 8
Command Openings 0
Max Tagged Openings 253
Device Queue Frozen Count 0
Channel A Target 1 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 2 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 3 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 4 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Goal: 3.300MB/s transfers
Curr: 3.300MB/s transfers
Channel A Target 4 Lun 0 Settings
Commands Queued 7193
Commands Active 1
Command Openings 0
Max Tagged Openings 0
Device Queue Frozen Count 0
Channel A Target 5 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Goal: 10.000MB/s transfers (10.000MHz, offset 15)
Curr: 10.000MB/s transfers (10.000MHz, offset 15)
Channel A Target 5 Lun 0 Settings
Commands Queued 854
Commands Active 0
Command Openings 253
Max Tagged Openings 253
Device Queue Frozen Count 0
Channel A Target 6 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Channel A Target 7 Negotiation Settings
User: 10.000MB/s transfers (10.000MHz, offset 255)
Thanks,
Karl <ko...@me...>
|
|
From: David M. B. <da...@me...> - 2001-08-14 03:46:17
|
Break a leg. Hey, feel free to nudge me into action. David "Karl O . Pinc" <ko...@me...> said: > I'm running torture tests on it now using the new 4GB drive in place of the > old 3GB drive. With any luck I won't have to crack the case again. > > Karl <ko...@me...> > > _______________________________________________ > hiccupmail-devel mailing list > hic...@li... > http://lists.sourceforge.net/lists/listinfo/hiccupmail-devel > -- David M. Bucknell http://members.iteachnet.org/~david/oldweb http://www.OpenSourceSchools.org International Education Daily http://members.iteachnet.org/webzine/ |
|
From: Karl O . P. <ko...@me...> - 2001-08-14 02:40:22
|
I'm running torture tests on it now using the new 4GB drive in place of the old 3GB drive. With any luck I won't have to crack the case again. Karl <ko...@me...> |
|
From: Karl O . P. <ko...@me...> - 2001-08-11 20:47:35
|
On 2001.08.09 10:28 Okwaro John wrote: > Hi Karl, > > How are you? Let me start by introducing myself. I am > John Okwaroh of National Museums of Kenya. Hi. Would you prefer John or Mr. Okwarch? Sorry to take so long to get back to you. > I was passed your letter by my director, Dr Goerge > Abungu. I have studied this issue you bruoght up and i > have found that it is a viable project which is > possible in Kenya. Great. > The reason why i am writing to you is we map out the > technical aspect on how we can carry out this. It can > be really interesting for so many institutions don't > have these facilities. I think it's interesting too. The general idea behind hiccupmail is twofold, first to provide a free service, and second to document both how to use the service and how to set up one like it. You can keep abrest of our progress by subscribing to the hic...@li... (or hic...@li...) mailing list. You can do this either via the web interface at http://lists.sourceforge.net/lists/listinfo/hiccupmail-devel or by sending a e-mail with a subject of "subscribe" (or "help") to the list's e-mail interface: hic...@li.... You may participate in the discussion or simply listen in. Here's the plan: Get the server hardware built. I hope to finish this weekend. (I've had an intermittent drive failure under load and it's been a pain.) Setup some infrastructure, like automated backup of the sourceforge.net development site, automated update of the hiccupmail.org web site from the deveopment site, and the like. Probably the beginnings of a Makefile for the project's code, maybe autoconf, and so forth. Then I will start putting software on the hiccupmail server. As I do so I'll document the process and the hiccupmail.org web page will begin to have instructions on how it's done. Once that's mostly done, I'll need to set up a "client" machine to test the server. This will also be documented on the web site. This is the point where you can begin to follow along in preparation for using the service. I expect to reach this point in about 2 months, but you'll have to keep tabs to see how it goes. (The next few weeks will see little progress as I'm traveling on business.) However, there is plenty you can do to prepare. Step one in preparing to use the server is probably going to be "Install RedHat Linux on your computer". This is a big step and not one that this project is prepared to provide too much help with at this point. There are other resources (the instruction manuals that come on the Linux CDs for example, or http://www.linuxdoc.org.) Certanly you won't need the full 2GB of Linux, or even a graphical user interface. I'll probably provide some sort of "follow along" instructions. And I'll likely start with a minimal linux install and add only a few packages. This will allow the client machines to save as much disk as possible for storing e-mail. By the way, although a stand-alone machine can be a hiccupmail client, you will get the most use of a machine attached to a LAN. That way the other machines on the LAN can be used to read/write the mail by many people. If you do not have a LAN, but have machines with modems and think that this is commonplace among potential hiccupmail users, the project could provide instructions so that people can get and send mail via modems. Anyhow, some minimal experience installing and using linux will be necessary on your part. The simple mechanics of using an unfamiliar text editor can be extremely frustrating, and as the client machines will be as low-end as possible, they won't have a grahpical interface and so some text editing will be necessary. (I recommend the emacs editor.) So, you can prepare by experimenting with Linux, or even simply browsing http://www.linuxdoc.org. I believe this will be worthwhile for you even in the absence of the hiccupmail project. I sent some Linux CDs through a roundabout route to your director but have had no word on their arrival. A few other notes: The buzz is that Redhat is close to a 7.2 release. I'd like to use the .2 releases as the basis of the project as these are most trouble free. I sent you the 7.1 release, and you probably won't have a problem. Also note that you can install Linux from the CDs on a machine with a LAN connection but no CD drive. You do this by putting the CDs into a machine on the LAN that is running a FTP server. (I'm sure a free ftp server for Windows can be found at tucows.com, although it may expire after a while.) I suspect that because Redhat Linux is now installed from 2 cds there is something more involved, but I don't know what. Perhaps the instructions are useful. Let's take any further discussion to the project's mailing list so all can benefit. (I'm cc-ing this message there.) If you're not familar with mailing lists, you should know that all list traffic is open to the public and is archived forever. So you might not want to say anything really embarrasing on the list about your mother-in-law or it might get back to her. ;-) Karl <ko...@me...> |
|
From: Karl O . P. <ko...@me...> - 2001-08-09 06:21:24
|
On 2001.08.08 23:49 David M. Bucknell wrote: > Sorry to hear about the problems; bright side is that anonymous friend. > Good going. I never heard from va Linux, by the way. PR takes time. We > can send that letter around some more, eh? At this point, I've about got hardware and have plenty to do. We're not in want of anything to progress far enough to have some "clients". I say hit them hard once we've proof of concept and try to get a second server to support more clients and for failover. Always better to ask for something really specific and have a pressing need. Karl <ko...@me...> |
|
From: David M. B. <da...@me...> - 2001-08-09 05:48:55
|
Sorry to hear about the problems; bright side is that anonymous friend. Good going. I never heard from va Linux, by the way. PR takes time. We can send that letter around some more, eh? David "Karl O . Pinc" <ko...@me...> said: > I'm still getting errors from the 3GB scsi drive. Near as I can figure > it's a seek failure. This could be due to an old power supply that just > can't provide the juice, it could be the drive needs a low-level format, or > it could be something else. I've re-arranged the drives in the bays and > souped-up the cooling and I don't think it's heat related because I've > gotten the error soon after power-up. > > The good news is that an anonymous donor supplied me with a modern (ish) > 4GB drive, so we can do without the old 3GB monster if need be. > > I'm going to make one more attempt to salvage the old drive and then move > on. > > Karl <ko...@me...> > > _______________________________________________ > hiccupmail-devel mailing list > hic...@li... > http://lists.sourceforge.net/lists/listinfo/hiccupmail-devel > -- David M. Bucknell http://members.iteachnet.org/~david/oldweb http://www.OpenSourceSchools.org International Education Daily http://members.iteachnet.org/webzine/ |
|
From: Karl O . P. <ko...@me...> - 2001-08-09 03:04:14
|
I'm still getting errors from the 3GB scsi drive. Near as I can figure it's a seek failure. This could be due to an old power supply that just can't provide the juice, it could be the drive needs a low-level format, or it could be something else. I've re-arranged the drives in the bays and souped-up the cooling and I don't think it's heat related because I've gotten the error soon after power-up. The good news is that an anonymous donor supplied me with a modern (ish) 4GB drive, so we can do without the old 3GB monster if need be. I'm going to make one more attempt to salvage the old drive and then move on. Karl <ko...@me...> |
|
From: David M. B. <da...@me...> - 2001-08-06 23:17:22
|
Heat related... Hmm. I thought all machines were supposed to be run with the panels off and the guts showing. Good luck. "Karl O . Pinc" <ko...@me...> said: > On 2001.08.06 02:25 Karl O . Pinc wrote: > > I've got the hardware assembled and am testing it under load for a few > > days > > to make sure it stands up. > > Got a few scsi errors overnight. I put a linux 2.4.7 on it (was 2.0.34) in > the hopes that would fix it. I've since gotten one error indicating a > drive problem. We'll see how it goes. It's probably heat releated. If it > keeps up I'll try running with the front panels off and see if that cools > better. > > Humm, the kernel seems to be making some automatic adjustments to the scsi > subsystem "(scsi0:A:0:0): Locking max tag count at 16" says the console. > Maybe I won't get any more errors. (Yeah, right.) > > Karl <ko...@me...> > > _______________________________________________ > hiccupmail-devel mailing list > hic...@li... > http://lists.sourceforge.net/lists/listinfo/hiccupmail-devel > -- David M. Bucknell http://members.iteachnet.org/~david/oldweb http://www.OpenSourceSchools.org International Education Daily http://members.iteachnet.org/webzine/ |
|
From: Karl O . P. <ko...@me...> - 2001-08-06 17:36:15
|
On 2001.08.06 02:25 Karl O . Pinc wrote: > I've got the hardware assembled and am testing it under load for a few > days > to make sure it stands up. Got a few scsi errors overnight. I put a linux 2.4.7 on it (was 2.0.34) in the hopes that would fix it. I've since gotten one error indicating a drive problem. We'll see how it goes. It's probably heat releated. If it keeps up I'll try running with the front panels off and see if that cools better. Humm, the kernel seems to be making some automatic adjustments to the scsi subsystem "(scsi0:A:0:0): Locking max tag count at 16" says the console. Maybe I won't get any more errors. (Yeah, right.) Karl <ko...@me...> |
|
From: David M. B. <da...@me...> - 2001-08-06 07:39:12
|
Great news. I'll put a gold star by your name for the week on the task list. DAvid "Karl O . Pinc" <ko...@me...> said: > I've got the hardware assembled and am testing it under load for a few days > to make sure it stands up. > > Development infrastructure is next. > > Karl <ko...@me...> > > _______________________________________________ > hiccupmail-devel mailing list > hic...@li... > http://lists.sourceforge.net/lists/listinfo/hiccupmail-devel > -- David M. Bucknell http://members.iteachnet.org/~david/oldweb http://www.OpenSourceSchools.org International Education Daily http://members.iteachnet.org/webzine/ |
|
From: Karl O . P. <ko...@me...> - 2001-08-06 07:19:31
|
I've got the hardware assembled and am testing it under load for a few days to make sure it stands up. Development infrastructure is next. Karl <ko...@me...> |
|
From: Karl O . P. <ko...@me...> - 2001-08-02 18:20:36
|
On 2001.08.02 01:53 David M. Bucknell wrote: > I created the task list. The SF tool allows you to choose tasks for > others so > I assigned everything to you, Karl ;-). > > I wasn't sure about priority ratings (it lets you choose colors -- ooh!) > and > dependencies got a bit foggy, too. Tell me what you don't like and I'll > change > it or feel free to change it/add items yourself. > > As we complete tasks, we should "report" them. Roger doger. Got the e-mails notifying me of the tasks. Neat. I'm hoping to have a master-off-sourceforge backup. I made a stab at putting that together but it dosen't work yet. I've made significant steps towards getting the hardware together. I want redundancy to help ensure we don't lose mail. I think we'll have 2GB of raid 1 (mirroring). However, it's in 2 1GB hunks, so we'll probably only have 1GB of queued delivery available at first. My random guess is that we can support between 10 and 30 clients with the hardware we have. Karl <ko...@me...> |
|
From: David M. B. <da...@me...> - 2001-08-02 07:52:01
|
I created the task list. The SF tool allows you to choose tasks for others so I assigned everything to you, Karl ;-). I wasn't sure about priority ratings (it lets you choose colors -- ooh!) and dependencies got a bit foggy, too. Tell me what you don't like and I'll change it or feel free to change it/add items yourself. As we complete tasks, we should "report" them. David -- David M. Bucknell http://members.iteachnet.org/~david/oldweb http://www.OpenSourceSchools.org International Education Daily http://members.iteachnet.org/webzine/ |
|
From: Karl O . P. <ko...@me...> - 2001-07-30 05:58:04
|
On 2001.07.29 20:13 David M. Bucknell wrote: > Hey Karl, > My sourceforge username is dbucknell."Karl O . Pinc" <ko...@me...> said: Ok. You've admin access. Please don't change anything on the live web site until I get CVS rocking. > > > On 2001.07.29 05:40 David M. Bucknell wrote: > > > Sorry, Karl. I can't see how to register myself as a member of the > > > hiccupmail project. Can you steer me to the right page? Karl <ko...@me...> |
|
From: David M. B. <da...@me...> - 2001-07-30 02:10:47
|
Hey Karl, My sourceforge username is dbucknell."Karl O . Pinc" <ko...@me...> said: > On 2001.07.29 05:40 David M. Bucknell wrote: > > Sorry, Karl. I can't see how to register myself as a member of the > > hiccupmail project. Can you steer me to the right page? > > I think I need to do it. But I need to know your SF unix name. > > > Also, tell me what job you want me to do. Is it task coordinator? > > That's good by me and would be very helpful. Tasks are: > > Figure how to use CVS as project admin. (kop) > > Establish development infrastructure: > autoconf > makefiles > see about literate programming with docbook > get website automatically updating from cvs > > Get first hiccupmail server up and document process. (kop) > > Get hardware for second server. > > Get bandwidth for second server. Ideally at a second physical location. > > Setup second server using hiccupmail procedures. > > Write setup procedure for client (kop?) > > Note: there are some task dependencies > > If you check out SF task management and find it would be easier for me to > add my own tasks let me know. If it's designed for everybody to managage > their own tasks we'll do it that way. In the mean time, you can do tasks. > Ok? Regardless, there's got to be somebody who coordinates and see's that > effort is not duplicated. I see you in that job. Make sense? Makes sense. Thanks. > What I really want is somebody to follow along with the instructions I > write to set up the server and see how good they are. Ideally this person > would be "joe average" and have the skillset and experience of someone > who's trying to set up their own hiccupmail service for the first time. > This review/feedback needn't be done as I write, but that might be best. > I'd like to produce our first secondary-hiccupmail-server that way, > although if necessay I'll do the work and review myself. Before setting up > a second server myself I want to write the howto for the client and get > some clients. I can be that guy both here and, if we're lucky, on another server. But if it's remote, someone else will have to start the installation, obviously. > I also want a great literary talent who will act as an editor for the > documentation. I just thought of a friend who's been a technical writer. > Perhaps he would help. It should be someone who will be willing to use the > CVS system so changes can be tracked. My friend may also have hardware and > be interested in constructing a hiccupmail server for us. I'll bcc him on > this (so his name dosen't hit the list.) Great! > My recommendation for the taskmaster is that if more than one person wants > to take on the same unassigned job, the parties involved should work out > themselves how to share. But you're it and can make your own policy. > > Karl <ko...@me...> > > _______________________________________________ > hiccupmail-devel mailing list > hic...@li... > http://lists.sourceforge.net/lists/listinfo/hiccupmail-devel > -- David M. Bucknell http://members.iteachnet.org/~david/oldweb http://www.OpenSourceSchools.org International Education Daily http://members.iteachnet.org/webzine/ |
|
From: Karl O . P. <ko...@me...> - 2001-07-29 21:58:39
|
I got this working yesterday. Karl <ko...@me...> |
|
From: Karl O . P. <ko...@me...> - 2001-07-29 20:36:31
|
This is good. I'd like to get it on the web site as a generic letter to potential donors. Can you make it html? Better yet, try typing it into "lyx" as a "letter" (IIRC). Then you can make ascii, html, or postscript if we ever want to print something really nice. Keep the lyx document as a master. Sound good? Before doing anything more to the web site, I'd like to figure out how to update it from CVS. It seems realy cool to keep a record of how the site has changed over time and should help us coordinate future changes. May as well start now. Are you willing to use entry level CVS? CVS is next task on the software side. Should I stop messing with hardware and get that setup so we can update the web site? On 2001.07.29 06:09 David M. Bucknell wrote: > Dear X, > > We are working on a project to help schools and other non profit orgs, > irrespective of geographical location, who may have difficulty > maintaining a > reliable e-mail presence. It's not the "reliable presence" schools have the a problem with, is it? >Karl Pinc, began the project > (registered on sourceforge) with the goal of enabling organizations which > have > intermittent Internet connections (e.g. modem dialup) to run their own > e-mail > servers; thereby giving them the freedom to make up their own e-mail > addresses. As a public service we plan to connect off-the-Internet e-mail > > servers to the Internet. This free service will be available to to > schools, > museums, and other worthy organizations on a first-come first-served > basis. We > will show you how we set up our service and help you provide the same > service > to anyone. > > The URL with relevant info is: http://www.hiccupmail.org > > We are looking for equipment and Internet co-loc and wonder if VA Linux > would > like to be involved. The project can use 2 relatively modern boxes in a > large > enclosure, with adequate cooling for plenty > expansion drives. We'd really like to do RAID 5 over RAID 0 and have bus > and > drive hardware scaled to support this RAID level in software. The project > will > recommend software raid as best practice to avoid being beholden to a particular hardware vendor for recovery of un-backed-up data after a controller failure. Our project does not currently have to resources to backup non-system data. The project has only one box presently and it is old (66Mhz 486!) > and > won't scale > very well. The plan for a second box is to replace the present box which > will > then be used to provide secondary MX and failover capability. A third > box is > required as a testbed when the project decides it's time to move to a > newer OS > version. We will also be looking for another co-location donor to > provide > geographic redundancy. We would, of course, be excited to have your help > in > these matters. > > As work on the Hiccupmail Project is done on a volunteer basis we make no > > warranties about > the speed of our progress or value of our products. We want to be clear > about > this so as not to make any mis-representations to potential donors. > Given the > project is its initial stages we understand if you feel you cannot > support us. > We will console ourselves by showing the world what old hardware is > capable of. Regardless of your decision we hope you remain receptive and > > receive with favor future requests we may make to you. > > Let us know if you have any suggestions for whom we might contact or ways > > satisfy our needs. > > Thank you. > > Sincerely, > > > > Karl O. Pinc -- hiccupmail.org > David Bucknell -- hiccupmail.org This is starting to look really good. Karl <ko...@me...> |
|
From: Karl O . P. <ko...@me...> - 2001-07-29 19:53:10
|
On 2001.07.29 05:40 David M. Bucknell wrote: > Sorry, Karl. I can't see how to register myself as a member of the > hiccupmail project. Can you steer me to the right page? I think I need to do it. But I need to know your SF unix name. > Also, tell me what job you want me to do. Is it task coordinator? That's good by me and would be very helpful. Tasks are: Figure how to use CVS as project admin. (kop) Establish development infrastructure: autoconf makefiles see about literate programming with docbook get website automatically updating from cvs Get first hiccupmail server up and document process. (kop) Get hardware for second server. Get bandwidth for second server. Ideally at a second physical location. Setup second server using hiccupmail procedures. Write setup procedure for client (kop?) Note: there are some task dependencies If you check out SF task management and find it would be easier for me to add my own tasks let me know. If it's designed for everybody to managage their own tasks we'll do it that way. In the mean time, you can do tasks. Ok? Regardless, there's got to be somebody who coordinates and see's that effort is not duplicated. I see you in that job. Make sense? What I really want is somebody to follow along with the instructions I write to set up the server and see how good they are. Ideally this person would be "joe average" and have the skillset and experience of someone who's trying to set up their own hiccupmail service for the first time. This review/feedback needn't be done as I write, but that might be best. I'd like to produce our first secondary-hiccupmail-server that way, although if necessay I'll do the work and review myself. Before setting up a second server myself I want to write the howto for the client and get some clients. I also want a great literary talent who will act as an editor for the documentation. I just thought of a friend who's been a technical writer. Perhaps he would help. It should be someone who will be willing to use the CVS system so changes can be tracked. My friend may also have hardware and be interested in constructing a hiccupmail server for us. I'll bcc him on this (so his name dosen't hit the list.) My recommendation for the taskmaster is that if more than one person wants to take on the same unassigned job, the parties involved should work out themselves how to share. But you're it and can make your own policy. Karl <ko...@me...> |
|
From: David M. B. <da...@me...> - 2001-07-29 12:06:38
|
Dear X, We are working on a project to help schools and other non profit orgs, irrespective of geographical location, who may have difficulty maintaining a reliable e-mail presence. Karl Pinc, began the project (registered on sourceforge) with the goal of enabling organizations which have intermittent Internet connections (e.g. modem dialup) to run their own e-mail servers; thereby giving them the freedom to make up their own e-mail addresses. As a public service we plan to connect off-the-Internet e-mail servers to the Internet. This free service will be available to to schools, museums, and other worthy organizations on a first-come first-served basis. We will show you how we set up our service and help you provide the same service to anyone. The URL with relevant info is: http://www.hiccupmail.org We are looking for equipment and Internet co-loc and wonder if VA Linux would like to be involved. The project can use 2 relatively modern boxes in a large enclosure, with adequate cooling for plenty expansion drives. We'd really like to do RAID 5 over RAID 0 and have bus and drive hardware scaled to support this RAID level in software. The project will recommend software raid as best avoids single-point-of-failure and so reduces hardware dependency which fosters long-term maintainability and minimal data loss. The project has only one box presently and it is old (66Mhz 486!) and won't scale very well. The plan for a second box is to replace the present box which will then be used to provide secondary MX and failover capability. A third box is required as a testbed when the project decides it's time to move to a newer OS version. We will also be looking for another co-location donor to provide geographic redundancy. We would, of course, be excited to have your help in these matters. As work on the Hiccupmail Project is done on a volunteer basis we make no warranties about the speed of our progress or value of our products. We want to be clear about this so as not to make any mis-representations to potential donors. Given the project is its initial stages we understand if you feel you cannot support us. We will console ourselves by showing the world what old hardware is capable of. Regardless of your decision we hope you remain receptive and receive with favor future requests we may make to you. Let us know if you have any suggestions for whom we might contact or ways satisfy our needs. Thank you. Sincerely, Karl O. Pinc -- hiccupmail.org David Bucknell -- hiccupmail.org -- David M. Bucknell http://members.iteachnet.org/~david/oldweb http://www.OpenSourceSchools.org International Education Daily http://members.iteachnet.org/webzine/ |
|
From: Karl O . P. <ko...@me...> - 2001-07-29 10:44:32
|
On 2001.07.29 02:48 David M. Bucknell wrote: > Here's what I wrote to clearlight.com > > Dear Don, > > I know you are sort of a family-style operation and that that's one > reason > you > do a good job. It may also be the reason you might not have the band > width to > get involved in the following, but any recommendations for possible > takers > would be gratefully received. > > I'm involved in a small project to set up a couple of e-mail servers to > accept > mail for schools that can't stay online all the time. Do you remember > the > NIST > account you set up for me in ...96 or 97? We had a local server but > dialed up > and fetched our mail and distributed it to local mailboxes. > > Here's the sourceforge application for this project: > > "The Hiccupmail Project is centered on providing a free mail relay > service > for > schools, museums, and other worthy organizations > which do not have a 24/7 net connection. The mail relay service itself > does > not involve SourceForge. SF's involvement will > primarily be as a central data repository and distribution point for the > > project and secondarily as a resource for the project's > code developers and documenters. The Hiccupmail Project's goal is to > enable > organizations with intermittent net connections to > run their own e-mail MTA and so the project will be distributing code > and > information to assist others who wish to replicate our > mail relay service. SourceForge will host a website for the project. SF > will > provide CVS and code distribution for any code > developed for the project. The first code to be developed will automate > the > sign-up process for users of the relay. " > > I'm wondering if you would be able to let us collocate a box at your > place > for > the project, or if you know of someone who might be interested and able > to > help and also to receive a little credit for a digital divide buster > project. > > Best wishes, > > David > > -- > David M. Bucknell > http://members.iteachnet.org/~david/oldweb > http://www.OpenSourceSchools.org > International Education Daily > http://members.iteachnet.org/webzine/ Thanks for the work. I'm not interested in pressing for more than a decent box to start with from anybody. And I've pretty specific requirements I'm looking for in a box. I feel our first priority now should be to get something working and that is not an inconsequential task. It needs some focus on my part. Both actual and visible progress is going to slow down at my end. Getting the hardware ready will consume some time and I've about used up my weekly allotment of free-time. I don't want to spend a lot of time corresponding, but some thought needs to go into the criteria we use to accept or reject an applicant for our free hiccupmail service. Do you want to try to organize some thoughts on this? My only thought is that we absolutely should not make a policy we can't enforce. Karl <ko...@me...> |
|
From: Karl O . P. <ko...@me...> - 2001-07-29 10:33:40
|
Whoops, I thought you were going to run it by me before sending it. I didn't realize it went out and worked on this copy. We can use this for later. Dear VA Linux, We are working on a project to help schools and other non profit orgs, irrespective of geographical location, who may have difficulty maintaining a reliable e-mail presence. Karl Pinc, began the project (registered on sourceforge) with the goal of enabling organizations which have intermittent Internet connections (e.g. modem dialup) to run their own e-mail servers; thereby giving them the freedom to make up their own e-mail addresses. As a public good we plan a free service to connect off-the-Internet e-mail servers to the Internet. This free service will be available to to schools, museums, and other worthy organizations on a first-come first-served basis. We will show you how we set up our service and help you provide the same service to anyone. The URL with relevant info is: http://www.hiccupmail.org We are looking for equipment and Internet co-loc and wonder if VA Linux would like to be involved. The project can use 2 relatively modern boxes in a large enclosure, with adequate cooling for plenty expansion drives. We'd really like to do raid 5 over raid 0 and have bus and drive hardware scaled to support this raid level in software. The project will recommend software raid as best avoids single-point-of-failure and so reduces hardware dependency which fosters long-term maintainability and minimal data loss. The project has only one box presently and it is old (66Mhz 486!) and won't scale very well. The plan for a second box is to replace the present box which will then be used to provide secondary MX and failover capability. A third box is required as a testbed when the project decides it's time to move to a newer OS version. We will also be looking for another co-location donor to provide geographic redundency. We would, of course, be excited to have your help in these matters. As work on the Hiccupmail Project is done on a volunteer basis we make no warrentees as to the speed of our progress or value of our products. We want to be clear about this so as not to make any mis-representations to potential donors. Given the progject is it's initial stages we understand if you feel you cannot support us. We will console ourselves by showing the world what old hardware is capable of. Regardless of your decision we hope you remain receptive and receive with favor future requests we may make to you. Let us know if you have any suggestions for whom we might contact or ways satisfy our needs. Thank you. Sincerely, David Bucknell -- hiccupmail.org, OpenSourceSchools.org, Iteachnet.org Karl O. Pinc -- hiccupmail.org -- |
|
From: David M. B. <da...@me...> - 2001-07-29 10:22:43
|
Re writing a how-to Have a look at http://www.riverdale.k12.or.us/linux/ They have some nice hot-to's that you might be able to build on or flatout recommend to potential hiccupmail users. As dictator I require you to forward this entire message to the list so we've a record and don't forget about it. I'm sure not going to remember. If you want to, start a todo page on the webserver or, better yet, be the (dum-de-dum-dum) taskmanager on sf.net. Looks easy. You set up a sub-project and add tasks when we need to make notes and remove them when we've accomplished them. You also assign them to developers. There's two of us and there's nothing wrong with keeping track of assignments and it can't hurt to be organized. That way we avoid overlapping effort and needless re-discussion of who should be asked about a task. Transparency is good. My tasklist item is to mention squirrelmail as a potential webmail program in our howto. Karl <ko...@me...> -- David M. Bucknell http://members.iteachnet.org/~david/oldweb http://www.OpenSourceSchools.org International Education Daily http://members.iteachnet.org/webzine/ |
|
From: David M. B. <da...@me...> - 2001-07-29 08:46:09
|
Here's what I wrote to clearlight.com Dear Don, I know you are sort of a family-style operation and that that's one reason you do a good job. It may also be the reason you might not have the band width to get involved in the following, but any recommendations for possible takers would be gratefully received. I'm involved in a small project to set up a couple of e-mail servers to accept mail for schools that can't stay online all the time. Do you remember the NIST account you set up for me in ...96 or 97? We had a local server but dialed up and fetched our mail and distributed it to local mailboxes. Here's the sourceforge application for this project: "The Hiccupmail Project is centered on providing a free mail relay service for schools, museums, and other worthy organizations which do not have a 24/7 net connection. The mail relay service itself does not involve SourceForge. SF's involvement will primarily be as a central data repository and distribution point for the project and secondarily as a resource for the project's code developers and documenters. The Hiccupmail Project's goal is to enable organizations with intermittent net connections to run their own e-mail MTA and so the project will be distributing code and information to assist others who wish to replicate our mail relay service. SourceForge will host a website for the project. SF will provide CVS and code distribution for any code developed for the project. The first code to be developed will automate the sign-up process for users of the relay. " I'm wondering if you would be able to let us collocate a box at your place for the project, or if you know of someone who might be interested and able to help and also to receive a little credit for a digital divide buster project. Best wishes, David -- David M. Bucknell http://members.iteachnet.org/~david/oldweb http://www.OpenSourceSchools.org International Education Daily http://members.iteachnet.org/webzine/ |
|
From: Karl O . P. <ko...@me...> - 2001-07-28 22:53:09
|
Hi, I've got a mailing list for the project: hic...@li... Let's move our discussions to the list. It'll be archived. I already put 2 of our mailings in the archive. There's also a home page http://hiccupmail.sourceforge.net. I'll eventually get around to making it www.hiccupmail.com so no links or bookmarks to it yet. It'd be good if you'd sign up to be a developer at sourceforge.net (sf.net). That way you could have access to edit the web page and otherwise work about the project. Ultimately, I'd like to produce the webpage with docbook. But not just now and we've lots of stuff to work out like mission statement and policy for who can use the free service. I think it's best posted publicly on the web page as it progresses. As project initiator I declare myself project-dictator. As project dictator I declare a commitment to full disclosure. You ok with this? You wanna be sub-dictator of any particular area? Please? Geeze, it's taken all day to set up the basics at SF. I know it will take at least that long to setup the CVS, autoconf, makefiles, etc. to begin the development. May as well begin to document setting up a server as I do the first one. This will require docbook so the docbook web page may come sooner rather than later. Karl <ko...@me...> |
