Menu
▾
▴
vacm-develop
You can subscribe to this list here.
| 2000 |
Jan
(2) |
Feb
(16) |
Mar
(66) |
Apr
(85) |
May
(50) |
Jun
(55) |
Jul
(44) |
Aug
(67) |
Sep
(27) |
Oct
(34) |
Nov
(34) |
Dec
(14) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2001 |
Jan
(26) |
Feb
(20) |
Mar
(28) |
Apr
(45) |
May
(5) |
Jun
(4) |
Jul
(2) |
Aug
|
Sep
(18) |
Oct
(9) |
Nov
(1) |
Dec
|
| 2002 |
Jan
|
Feb
|
Mar
|
Apr
(1) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2007 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(1) |
Nov
|
Dec
|
|
From: Rob L. <rl...@pl...> - 2001-09-13 18:11:55
|
On Thu, Sep 13, 2001 at 11:02:46AM -0700, San Mehat wrote: > There is a 2.4 driver.. ill dig it up Ok, so i found one in cvs for impitools. But in http://www.geocrawler.com/archives/3/1160/2001/5/0/5749716/ you say this patch causes "undesired behavior" , and that a new one is in the works. All the timestamps in the cvs patch are january 26, and since your message is mid-may, i'm hoping there's a working patch "somewhere" :> if you could post that patch when you find it, that'd be great. or just clarify what "undesired behavior" means, and maybe i can live with that :> ==rob -- [ Rob Latham <rl...@pl...> Developer, Admin, Alchemist ] [ Paralogic Inc. - www.plogic.com ] [ ] [ EAE8 DE90 85BB 526F 3181 1FCF 51C4 B6CB 08CC 0897 ] |
|
From: San M. <net...@va...> - 2001-09-13 18:02:28
|
Heh thanks.. hopefully the vacm-develop mailing list wont turn into a 'help-find-san-a-job' list ;) -----Original Message----- From: vac...@li... [mailto:vac...@li...] On Behalf Of Rob Latham Sent: Thursday, September 13, 2001 11:01 AM To: San Mehat Cc: vac...@li... Subject: Re: [Vacm-develop] IPMI and 2.4 kernels On Thu, Sep 13, 2001 at 11:02:46AM -0700, San Mehat wrote: > Well I'm still here.. lol... although not in software engineering > anymore... I haven't decided what to do about vacm... its difficult to > get hardware specs when you're just a lone guy doing the project.. but > we'll see... > I need to figure it out... among a zillion other things.... > > There is a 2.4 driver.. ill dig it up I found one that seems to work. at least, it applies w/o any rejects http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/~checkout~/ipmitools/ipmi tools/kernel/kcs/patches/2.4.x/2.4.0.ipmi_kcs_2.0.patcn Thanks. and i'm glad to hear VA was able to keep at least a few quality people. ==rob -- [ Rob Latham <rl...@pl...> Developer, Admin, Alchemist ] [ Paralogic Inc. - www.plogic.com ] [ ] [ EAE8 DE90 85BB 526F 3181 1FCF 51C4 B6CB 08CC 0897 ] _______________________________________________ Vacm-develop mailing list Vac...@li... https://lists.sourceforge.net/lists/listinfo/vacm-develop |
|
From: Rob L. <rl...@pl...> - 2001-09-13 18:00:37
|
On Thu, Sep 13, 2001 at 11:02:46AM -0700, San Mehat wrote: > Well I'm still here.. lol... although not in software engineering > anymore... I haven't decided what to do about vacm... its difficult to > get hardware specs when you're just a lone guy doing the project.. but > we'll see... > I need to figure it out... among a zillion other things.... > > There is a 2.4 driver.. ill dig it up I found one that seems to work. at least, it applies w/o any rejects http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/~checkout~/ipmitools/ipmitools/kernel/kcs/patches/2.4.x/2.4.0.ipmi_kcs_2.0.patcn Thanks. and i'm glad to hear VA was able to keep at least a few quality people. ==rob -- [ Rob Latham <rl...@pl...> Developer, Admin, Alchemist ] [ Paralogic Inc. - www.plogic.com ] [ ] [ EAE8 DE90 85BB 526F 3181 1FCF 51C4 B6CB 08CC 0897 ] |
|
From: San M. <net...@va...> - 2001-09-13 17:48:38
|
Well I'm still here.. lol... although not in software engineering anymore... I haven't decided what to do about vacm... its difficult to get hardware specs when you're just a lone guy doing the project.. but we'll see... I need to figure it out... among a zillion other things.... There is a 2.4 driver.. ill dig it up -----Original Message----- From: vac...@li... [mailto:vac...@li...] On Behalf Of Rob Latham Sent: Thursday, September 13, 2001 10:46 AM To: vac...@li... Subject: [Vacm-develop] IPMI and 2.4 kernels I'd like to use vacm and it's IPMI support on a 2.4 kernel: any idea if a: whoever at va was working on IPMI support for 2.2 is still employed there or b: if anyone elsewhere has done IMPI support for the linux kernel ? thanks ==rob -- [ Rob Latham <rl...@pl...> Developer, Admin, Alchemist ] [ Paralogic Inc. - www.plogic.com ] [ ] [ EAE8 DE90 85BB 526F 3181 1FCF 51C4 B6CB 08CC 0897 ] _______________________________________________ Vacm-develop mailing list Vac...@li... https://lists.sourceforge.net/lists/listinfo/vacm-develop |
|
From: Rob L. <rl...@pl...> - 2001-09-13 17:45:47
|
I'd like to use vacm and it's IPMI support on a 2.4 kernel: any idea if a: whoever at va was working on IPMI support for 2.2 is still employed there or b: if anyone elsewhere has done IMPI support for the linux kernel ? thanks ==rob -- [ Rob Latham <rl...@pl...> Developer, Admin, Alchemist ] [ Paralogic Inc. - www.plogic.com ] [ ] [ EAE8 DE90 85BB 526F 3181 1FCF 51C4 B6CB 08CC 0897 ] |
|
From: Adam L. <ad...@la...> - 2001-07-26 22:39:42
|
I have vacm 2.0.5 running on a small cluster of 4 rhat 7.1 machines. I downloaded and got vacm-perl to compile and install, but I get segfaults when running very simple vacm-perl code. Is vacm-perl compatible with vacm 2.0.5? Are they both functional on redhat 7.1? -- Adam Lazur, Cluster Monkey |
|
From: John G. <jg...@va...> - 2001-07-25 03:39:11
|
San, et al, I have a large APC that I would like to use with VACM/APC module. Is it ready to use in a productions system? Also, is there any dependencies on the CVS version of VACM that APC needs; I have a client that is using 2.0.5. Thanks for the help, John John Goebel<jg...@va...> VA Linux |
|
From: <sg...@va...> - 2001-06-08 00:26:10
|
On Thu, 7 Jun 2001, Dale Harris wrote: > Personally, I pretty much always put a: > > alias char-major-46 rocket > > in /etc/conf.modules (/etc/modules.conf, if you are using anything other > than Red Hat) /etc/rc.d/init.d/vacm checks for the need for a rocket module (/etc/sysconfig/vacm; ROCKET=yes) and loads if needed. The problem was the config file wasn't set correctly. My point was that a better error message would have directed my investigations towards the config file, not towards a possible problem in VACM itself. (This needs to be in support's troubleshooting hints and kinks) Steve ------------------------------------------------------------------------- Stephan A. Greene sg...@va... Senior Systems Specialist, VA Linux Systems 1-888-LINUX-4U x76774 Mid-Atlantic Region 1-703-980-3475 Amateur KA...@am... Grid FM18hx 38 59'83.33"N 77 23'6.15"W ------------------------------------------------------------------------- |
|
From: Dale H. <ro...@va...> - 2001-06-08 00:04:07
|
On Thu, Jun 07, 2001 at 05:41:25PM -0400, Stephan Greene wrote: > The underlying problem is the error messages returned by sercon, emp, and > other modules. Rather than saying "the command failed becuase I'm > configured to use a serial port that doesn't exist on this system", they > report "the command failed becuase I'm not really loaded even though I say > I am". > Personally, I pretty much always put a: alias char-major-46 rocket in /etc/conf.modules (/etc/modules.conf, if you are using anything other than Red Hat) -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Dale Harris <ro...@va...> VA Linux Systems Senior Support Engineer 47071 Bayview Pkwy. (877) VA-LINUX Fremont, CA 94538 |
|
From: San 'N. M. <net...@va...> - 2001-06-07 22:15:33
|
Stephan,
SERCON reports what the kernel tells us when we try to open() the device.
The kernel treats a driver not loaded the same as if the hardware didn't
exist.. so its not really possibly for an application to figure out *why*
the open failed without doing more device specific snooping.. which is not
really supported by the linux character device system..
here is the code in question:
if ((ctx->con_sock = open(ctx->device_address, O_RDWR)) < 0)
{
sprintf(log_msg,
"0:0:LOG:Thread %s engage failed (%m)",
ctx->node_id);
send_msg(ctx->thread_ipc_write, log_msg);
return(-errno);
}
as you can see... if we cant open the device, we report an error.. the (%m)
is a string created from the return code the kernel sends us...
so its not really vacm reporting useless error messages, but more the kernel
doing it..
-san
The error messages
----- Original Message -----
From: "Stephan Greene" <sg...@va...>
To: "VACM Development List" <vac...@li...>
Sent: Thursday, June 07, 2001 2:41 PM
Subject: [Vacm-develop] Vacm useless error message behavior
>
> I was working on a cluster today that uses rocketport cards for emp and
> serial access. Whenever I tried running the emp or sercon modules from
> within vash, I would get an error about the module not being loaded or
> somesuch. Meanwhile, the appropriate nexxus command showed EMP and SERCON
> *were* loaded.
>
> lsmod showed rocket.o was NOT loaded. Strange. But a manual "insmod
> rocket" fixed the problem. Yet the vacm startup script clearly checks for
> the rocketports and loads them if needed.
>
> The problem turned out to be "ROCKET=false" in /etc/sysconfig/vacm. This
> stopped the loading of the module. Once this was fixed, so was the
> problem.
>
> The underlying problem is the error messages returned by sercon, emp, and
> other modules. Rather than saying "the command failed becuase I'm
> configured to use a serial port that doesn't exist on this system", they
> report "the command failed becuase I'm not really loaded even though I say
> I am".
>
> If possible, a better error message would speed debugging configurations
> in the field.
>
> Thanks
>
> Steve
>
> -------------------------------------------------------------------------
> Stephan A. Greene sg...@va...
> Senior Systems Specialist, VA Linux Systems 1-888-LINUX-4U x76774
> Mid-Atlantic Region 1-703-980-3475
> Amateur KA...@am... Grid FM18hx 38 59'83.33"N 77 23'6.15"W
> -------------------------------------------------------------------------
>
>
> _______________________________________________
> Vacm-develop mailing list
> Vac...@li...
> http://lists.sourceforge.net/lists/listinfo/vacm-develop
>
|
|
From: Stephan G. <sg...@va...> - 2001-06-07 21:41:27
|
I was working on a cluster today that uses rocketport cards for emp and serial access. Whenever I tried running the emp or sercon modules from within vash, I would get an error about the module not being loaded or somesuch. Meanwhile, the appropriate nexxus command showed EMP and SERCON *were* loaded. lsmod showed rocket.o was NOT loaded. Strange. But a manual "insmod rocket" fixed the problem. Yet the vacm startup script clearly checks for the rocketports and loads them if needed. The problem turned out to be "ROCKET=false" in /etc/sysconfig/vacm. This stopped the loading of the module. Once this was fixed, so was the problem. The underlying problem is the error messages returned by sercon, emp, and other modules. Rather than saying "the command failed becuase I'm configured to use a serial port that doesn't exist on this system", they report "the command failed becuase I'm not really loaded even though I say I am". If possible, a better error message would speed debugging configurations in the field. Thanks Steve ------------------------------------------------------------------------- Stephan A. Greene sg...@va... Senior Systems Specialist, VA Linux Systems 1-888-LINUX-4U x76774 Mid-Atlantic Region 1-703-980-3475 Amateur KA...@am... Grid FM18hx 38 59'83.33"N 77 23'6.15"W ------------------------------------------------------------------------- |
|
From:
<and...@vi...> - 2001-05-29 12:35:02
|
Hi Developers. I have recently installed a server with RedHat 7.1 and compiled and installed a kernel 2.4.4 I cannot get VACM to work properly. When i try to connect to nexxus with vash, I get the following error-message: [root@netman /root]# vash -c localhost vash: Unable to ping nexxus at localhost vash$ If I use the kernel, that comes with RedHat 7.1 as standard, it connects in approx. 80% of the times, but with my own compilation its around 5% of the times or less. I may have a problem with my new kernel, but what am I supposed to be careful about, when I configure, that causes these kind of problems? Kind regards, Anders >>>> VIGILANTe.com NOTICE - AUTOMATICALLY INSERTED <<<< The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. Any opinions expressed in this email are those of the individual and not necessarily the Company. If you receive this transmission in error, please email to pos...@vi..., including a copy of this message. Please then delete this email and destroy any copies of it. >>>>>>>>>>>>>>>>>>>>>>>>>> DISCLAIMER END <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< |
|
From: Adam M. <ma...@si...> - 2001-05-14 16:53:28
|
I have been experiencing problems with the vacm 2.0.5 EMP module sending
repeated commands. I setup up a quick test to verify that the EMP module
would be reliable enough for use with GFS. The test is pretty simple:
while sleep 60
do
vash -c nexxus -u user -p pass -x "ipc nexxus emp:power_cycle:node"
done
After between two and six hours, the EMP commands fail. For some reason,
the emp BIOS values get reset and must be reconfigured in order to
make vacm emp work again.
Do you have any idea why the BIOS values would be reset?
The following is the bios information that is displayed on boot:
Copyright 1985-2000 Phoenix Technologies Ltd., All Rights Reserved.
L440GX0.86B.0125.P13.0006061627
L440GX+ Production Release 13.0
BIOS Build 125
The following is the settings in the bios pretaining to EMP:
Server Management Info
System Event Logging: [Enabled]
Clear Event Log: [No]
EMP Password Switch: [Disabled]
EMP ESC Sequence [+++ ]
EMP Hangup Line String: [ATH ]
Modem Init String: [AT&F0S0=1S14=0&D]
High Modem Init String: [0 ]
EMP Access Mode [Always Active]
EMP Restricted Mode Access [Disabled]
EMP Direct Connect/Modem Mode [Direct Connect]
--
Adam Manthei <ma...@si...>
|
|
From: San M. <net...@va...> - 2001-05-14 15:50:29
|
Hey Greg, the ipmitools project contains patches for 2.2.x non va kernels... the 2.4.x patches that are in the ipmitools project cvs tree should not be applied as they are known to cause undesired behavior.... we are currently qualifying a new 2.4 patch which will be included in the source asap -san Greg Crosby wrote: > Hi > > we are currently testing vacm we have built a test rig from valinux equip > 2* 1120's > 1* 2240 (vacm management node) > 1*nas 9205 > > but because our current live servers on-site are running mandrake we have installed mandrake 7.2 on one of the 1120's and mandrake 8.0 (which we are currently evaluating....fast tcp/ip and stateful firewall) on the other 1120 machine > > now emp_extentd obviously won't load because there is no ipmi_kcs support in the ,mandrake kernels (2.2.18 and 2.4.3 respectively) > > are there patches which will work for patching the kernel source of a non valinux kernel to include the ipmi_kcs support, firstly for mdk 7.2 and what stage are you at with 2.4.x. > > cheers! > > Greg Crosby > Pre-Sales Consultant > Netkonect Communications plc > Opus House, Manor Court, Herriard, Basingstoke, RG25 2PH > > T: 0870 063 2222 F: 0870 063 4444 > W: www.netkonect.net E: gc...@ne... > > BUILDING YOUR BUSINESS CONNECTIONS > > This e-mail including any attachments is confidential and may be legally > privileged. If you have received it in error please advise the sender > immediately by return email and then delete it from your system. The > unauthorised use, distribution, copying or alteration of this email is > strictly forbidden. If you need assistance please contact the Netkonect > helpdesk on 0870 063 2222. > > _______________________________________________ > Vacm-develop mailing list > Vac...@li... > http://lists.sourceforge.net/lists/listinfo/vacm-develop |
|
From: Greg C. <gc...@ne...> - 2001-05-14 08:56:16
|
Hi we are currently testing vacm we have built a test rig from valinux equip 2* 1120's 1* 2240 (vacm management node) 1*nas 9205 but because our current live servers on-site are running mandrake we have = installed mandrake 7.2 on one of the 1120's and mandrake 8.0 (which we are = currently evaluating....fast tcp/ip and stateful firewall) on the other 1120 = machine now emp_extentd obviously won't load because there is no ipmi_kcs support in = the ,mandrake kernels (2.2.18 and 2.4.3 respectively) are there patches which will work for patching the kernel source of a non = valinux kernel to include the ipmi_kcs support, firstly for mdk 7.2 and what = stage are you at with 2.4.x. cheers! Greg Crosby Pre-Sales Consultant Netkonect Communications plc Opus House, Manor Court, Herriard, Basingstoke, RG25 2PH T: 0870 063 2222 F: 0870 063 4444 W: www.netkonect.net E: gc...@ne... BUILDING YOUR BUSINESS CONNECTIONS This e-mail including any attachments is confidential and may be legally privileged. If you have received it in error please advise the sender immediately by return email and then delete it from your system. The unauthorised use, distribution, copying or alteration of this email is strictly forbidden. If you need assistance please contact the Netkonect helpdesk on 0870 063 2222. |
|
From: Greg C. <gc...@ne...> - 2001-05-14 08:37:08
|
confirm 658348 |
|
From: Dirk W. <di...@re...> - 2001-04-28 17:47:39
|
On Sat, 28 Apr 2001, Wichert Akkerman wrote: > Previously Wichert Akkerman wrote: > > There is http://sourceforge.net/projects/ipmitools/ which as a .tar.gz > > with ipmi_ctl though, but indeed no patch. > > Actually there is a patch, but it's hidden in the CVS tree for > the ipmitools projects. great, you're right. here it is: cvs.sourceforge.net/cgi-bin/viewcvs.cgi/ipmitools/ipmitools/kernel/kcs/patches/\ 2.4.x/ ~dirkw |
|
From: Wichert A. <wi...@ci...> - 2001-04-28 17:42:42
|
Previously Wichert Akkerman wrote: > There is http://sourceforge.net/projects/ipmitools/ which as a .tar.gz > with ipmi_ctl though, but indeed no patch. Actually there is a patch, but it's hidden in the CVS tree for the ipmitools projects. Wichert. -- ________________________________________________________________ / Generally uninteresting signature - ignore at your convenience \ | wi...@ci... http://www.liacs.nl/~wichert/ | | 1024D/2FA3BC2D 576E 100B 518D 2F16 36B0 2805 3CB8 9250 2FA3 BC2D | |
|
From: Wichert A. <wi...@ci...> - 2001-04-28 17:40:17
|
Previously Dirk Wetter wrote: > - snip - > Welcome to http://ipmitools.sourceforge.net/ There is http://sourceforge.net/projects/ipmitools/ which as a .tar.gz with ipmi_ctl though, but indeed no patch. Wichert. -- ________________________________________________________________ / Generally uninteresting signature - ignore at your convenience \ | wi...@ci... http://www.liacs.nl/~wichert/ | | 1024D/2FA3BC2D 576E 100B 518D 2F16 36B0 2805 3CB8 9250 2FA3 BC2D | |
|
From: Dirk W. <di...@re...> - 2001-04-28 17:08:33
|
On Sun, 29 Apr 2001, Yoshiaki Yanagihara wrote: > > On Sat, 28 Apr 2001 02:11:03 +0200 > Wichert Akkerman <wi...@va...> wrote: > > > > > The documentation mentions the IPMI patch can be found in the > > nexxus/nexxus_modules/emp/support_utilities/ subdirectory, > > but it appears the patch is not present anywhere in the source. > > Since no other location is given it seems I'm a bit stuck. > > > > It also seems there still is no IPMI patch for 2.4.*, is that > > correct? > > The latest IPMI patch for 2.2.x and 2.4.x is included in the ipmitools > (http://ipmitools.sourceforge.net/). this is what i get, if i go to ipmitools.sourceforge.net: - snip - Welcome to http://ipmitools.sourceforge.net/ We're Sorry but this Project hasn't yet uploaded their personal webpage yet. Please check back soon for updates or visit SourceForge - snap - ~dirkw |
|
From: Yoshiaki Y. <yo...@va...> - 2001-04-28 16:56:29
|
On Sat, 28 Apr 2001 02:11:03 +0200 Wichert Akkerman <wi...@va...> wrote: > > The documentation mentions the IPMI patch can be found in the > nexxus/nexxus_modules/emp/support_utilities/ subdirectory, > but it appears the patch is not present anywhere in the source. > Since no other location is given it seems I'm a bit stuck. > > It also seems there still is no IPMI patch for 2.4.*, is that > correct? The latest IPMI patch for 2.2.x and 2.4.x is included in the ipmitools (http://ipmitools.sourceforge.net/). Regards, Yochi -- Yoshiaki Yanagihara <yo...@va...> <yo...@de...> |
|
From: Wichert A. <wi...@ci...> - 2001-04-28 00:49:27
|
In an effort to not appear as a lazy bum I cobbled up ipmi-kcs 2.0 patch for the 2.2.19 kernel. Expect it to show up in Debian in the next 48 hours. Wichert. -- ________________________________________________________________ / Generally uninteresting signature - ignore at your convenience \ | wi...@ci... http://www.liacs.nl/~wichert/ | | 1024D/2FA3BC2D 576E 100B 518D 2F16 36B0 2805 3CB8 9250 2FA3 BC2D | |
|
From: Dirk W. <di...@re...> - 2001-04-28 00:14:41
|
On Sat, 28 Apr 2001, Wichert Akkerman wrote: > > The documentation mentions the IPMI patch can be found in the > nexxus/nexxus_modules/emp/support_utilities/ subdirectory, > but it appears the patch is not present anywhere in the source. > Since no other location is given it seems I'm a bit stuck. > > It also seems there still is no IPMI patch for 2.4.*, is that > correct? i am looking for that patch, too ;-( ~dirkw |
|
From: Wichert A. <wi...@va...> - 2001-04-28 00:11:10
|
The documentation mentions the IPMI patch can be found in the nexxus/nexxus_modules/emp/support_utilities/ subdirectory, but it appears the patch is not present anywhere in the source. Since no other location is given it seems I'm a bit stuck. It also seems there still is no IPMI patch for 2.4.*, is that correct? Wichert. -- ________________________________________________________________ / Generally uninteresting signature - ignore at your convenience \ | wi...@ci... http://www.liacs.nl/~wichert/ | | 1024D/2FA3BC2D 576E 100B 518D 2F16 36B0 2805 3CB8 9250 2FA3 BC2D | |
|
From: Wichert A. <wi...@va...> - 2001-04-27 23:44:07
|
Previously ra...@ra... wrote: > glade builds the entire flim makefile configure etc. stuff. its all > automake./autoconf built. those paths there are just build from the > stuff theat automake. autoconf and gettext spew out - all of whihc are > gnu tools.. Oh great, more build-dependencies. Wouldn't it make sense to split flim and hoover into their own .tar.gz so things are a bit simpler? The amount of buildtime interdepencies you get now is somewhat nasty. > make distcheck works and making with a buildroot works :) to be honest > - i'm not sure why you do it that way.. its odd. I like having my objects in a different directory from my sources. Makes it very easy to clean my buildtree and spot errors in the build process. Wichert. -- ________________________________________________________________ / Generally uninteresting signature - ignore at your convenience \ | wi...@ci... http://www.liacs.nl/~wichert/ | | 1024D/2FA3BC2D 576E 100B 518D 2F16 36B0 2805 3CB8 9250 2FA3 BC2D | |
