Menu
▾
▴
moosefs-users — Mailing list for MooseFS Users
You can subscribe to this list here.
| 2009 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(4) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2010 |
Jan
(20) |
Feb
(11) |
Mar
(11) |
Apr
(9) |
May
(22) |
Jun
(85) |
Jul
(94) |
Aug
(80) |
Sep
(72) |
Oct
(64) |
Nov
(69) |
Dec
(89) |
| 2011 |
Jan
(72) |
Feb
(109) |
Mar
(116) |
Apr
(117) |
May
(117) |
Jun
(102) |
Jul
(91) |
Aug
(72) |
Sep
(51) |
Oct
(41) |
Nov
(55) |
Dec
(74) |
| 2012 |
Jan
(45) |
Feb
(77) |
Mar
(99) |
Apr
(113) |
May
(132) |
Jun
(75) |
Jul
(70) |
Aug
(58) |
Sep
(58) |
Oct
(37) |
Nov
(51) |
Dec
(15) |
| 2013 |
Jan
(28) |
Feb
(16) |
Mar
(25) |
Apr
(38) |
May
(23) |
Jun
(39) |
Jul
(42) |
Aug
(19) |
Sep
(41) |
Oct
(31) |
Nov
(18) |
Dec
(18) |
| 2014 |
Jan
(17) |
Feb
(19) |
Mar
(39) |
Apr
(16) |
May
(10) |
Jun
(13) |
Jul
(17) |
Aug
(13) |
Sep
(8) |
Oct
(53) |
Nov
(23) |
Dec
(7) |
| 2015 |
Jan
(35) |
Feb
(13) |
Mar
(14) |
Apr
(56) |
May
(8) |
Jun
(18) |
Jul
(26) |
Aug
(33) |
Sep
(40) |
Oct
(37) |
Nov
(24) |
Dec
(20) |
| 2016 |
Jan
(38) |
Feb
(20) |
Mar
(25) |
Apr
(14) |
May
(6) |
Jun
(36) |
Jul
(27) |
Aug
(19) |
Sep
(36) |
Oct
(24) |
Nov
(15) |
Dec
(16) |
| 2017 |
Jan
(8) |
Feb
(13) |
Mar
(17) |
Apr
(20) |
May
(28) |
Jun
(10) |
Jul
(20) |
Aug
(3) |
Sep
(18) |
Oct
(8) |
Nov
|
Dec
(5) |
| 2018 |
Jan
(15) |
Feb
(9) |
Mar
(12) |
Apr
(7) |
May
(123) |
Jun
(41) |
Jul
|
Aug
(14) |
Sep
|
Oct
(15) |
Nov
|
Dec
(7) |
| 2019 |
Jan
(2) |
Feb
(9) |
Mar
(2) |
Apr
(9) |
May
|
Jun
|
Jul
(2) |
Aug
|
Sep
(6) |
Oct
(1) |
Nov
(12) |
Dec
(2) |
| 2020 |
Jan
(2) |
Feb
|
Mar
|
Apr
(3) |
May
|
Jun
(4) |
Jul
(4) |
Aug
(1) |
Sep
(18) |
Oct
(2) |
Nov
|
Dec
|
| 2021 |
Jan
|
Feb
(3) |
Mar
|
Apr
|
May
|
Jun
|
Jul
(6) |
Aug
|
Sep
(5) |
Oct
(5) |
Nov
(3) |
Dec
|
| 2022 |
Jan
|
Feb
|
Mar
(3) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Michał B. <mic...@ge...> - 2010-05-12 17:42:00
|
You have lots of chunks on your chunkservers. Probably the registration packet for the master is too big and the master server rejects it. You need to change in the "matocsserv.c" file in "mfsmaster" folder this line: #define MaxPacketSize 50000000 into this: #define MaxPacketSize 200000000 After changing source you need to recompile the master server and restart it. Generally on one chunkserver there are about 500,000 chunks. For this amount it is necessary to allocate 6 million bytes. Limit of 50 million seemed quite reasonable. But you have about 5,000,000 chunks on one chunkserver which demands about 60 million bytes for sending information about these chunks. And you exceed this limit. That's why we suggest to increase it to 200 million bytes. Let us know if it fixes your problem. Kind regards Michał Borychowski From: lwxian_aha [mailto:lwx...@16...] Sent: Tuesday, May 11, 2010 5:34 AM To: moosefs-users Subject: [Moosefs-users] chunkserver can't connect to masterserver today I have new trouble with my new MFS system,chunkserver can't connect to masterserver; my MFS system consist of one masterserver,three chunkserver,every chunkserver with 7.2T diskspace; about 1.4T data and about 7 million files ,every file with 2 copies; MFS version is 1.6.11 OS version CENTOS 5.0 FS is ext3; Following is the error message: [root@localhost mfs]# tail /var/log/messages May 11 10:57:00 localhost mfsmaster[25802]: server 1 (ip: 192.168.10.23, port: 9422): usedspace: 33326616576 (31.04 GiB), totalspace: 140025790464 (130.41 GiB), usage: 23.80% May 11 10:57:00 localhost mfsmaster[25802]: total: usedspace: 33326616576 (31.04 GiB), totalspace: 140025790464 (130.41 GiB), usage: 23.80% May 11 10:57:03 localhost mfsmaster[25802]: CS(192.168.10.21) packet too long (56130653/50000000) May 11 10:57:03 localhost mfsmaster[25802]: chunkserver disconnected - ip: 192.168.10.21, port: 0, usedspace: 0 (0.00 GiB), totalspace: 0 (0.00 GiB) May 11 10:57:08 localhost mfsmaster[25802]: CS(192.168.10.21) packet too long (56130653/50000000) May 11 10:57:08 localhost mfsmaster[25802]: chunkserver disconnected - ip: 192.168.10.21, port: 0, usedspace: 0 (0.00 GiB), totalspace: 0 (0.00 GiB) May 11 10:57:13 localhost mfsmaster[25802]: CS(192.168.10.21) packet too long (56130653/50000000) May 11 10:57:13 localhost mfsmaster[25802]: chunkserver disconnected - ip: 192.168.10.21, port: 0, usedspace: 0 (0.00 GiB), totalspace: 0 (0.00 GiB) May 11 10:57:18 localhost mfsmaster[25802]: CS(192.168.10.21) packet too long (56130653/50000000) May 11 10:57:18 localhost mfsmaster[25802]: chunkserver disconnected - ip: 192.168.10.21, port: 0, usedspace: 0 (0.00 GiB), totalspace: 0 (0.00 GiB) what's happen ?need you help! thanks a lot's 2010-05-05 ________________________________________ lwxian_aha |
|
From: jose m. <aso...@zo...> - 2010-05-12 17:22:34
|
El Lunes 10 Mayo 2010, lwxian_aha escribió:
> 158922: photos/47/62/63/m_853671.jpg May 10 20:34:58 localhost
> mfsmaster[20168]: currently unavailable chunk 00000000004465CB (inode:
> 4353227 ; index: 0) May 10 20:34:58 localhost mfsmaster[20168]: *
> currently unavailable file 4353227: photos/46/45/10/s_848116.jpg May 10
> 20:34:58 localhost mfsmaster[20168]: currently unavailable chunk
> 0000000000026B8E (inode: 158923 ; index: 0) May 10 20:34:58 localhost
> mfsmaster[20168]: * currently unavailable file 158923:
> photos/47/62/63/739789.jpg May 10 20:34:58 localhost mfsmaster[20168]:
> currently unavailable chunk 00000000004465CC (inode: 4353228 ; index: 0)
> May 10 20:34:58 localhost mfsmaster[20168]: * currently unavailable file
> 4353228: photos/46/91/73/936449.jpg May 10 20:34:58 localhost
> mfsmaster[20168]: currently unavailable chunk 00000000000CBC48 (inode:
> 158924 ; index: 0) May 10 20:34:58 localhost mfsmaster[20168]: * currently
> unavailable file 158924: photos/47/74/82/s_171883.jpg May 10 20:34:58
> localhost mfsmaster[20168]: currently unavailable chunk 00000000004465CD
> (inode: 4353229 ; index: 0) May 10 20:34:58 localhost mfsmaster[20168]: *
> currently unavailable file 4353229: photos/46/45/10/l_848116.jpg May 10
> 20:34:58 localhost mfsmaster[20168]: currently unavailable chunk
> 00000000000CBC49 (inode: 158925 ; index: 0) May 10 20:34:58 localhost
> mfsmaster[20168]: * currently unavailable file 158925:
> photos/47/74/82/m_156158.jpg May 10 20:34:58 localhost mfsmaster[20168]:
> currently unavailable chunk 00000000004465CE (inode: 4353230 ; index: 0)
> May 10 20:34:58 localhost mfsmaster[20168]: * currently unavailable file
> 4353230: photos/46/91/73/m_928481.jpg May 10 20:34:58 localhost
> mfsmaster[20168]: currently unavailable chunk 0000000000026B91 (inode:
> 158926 ; index: 0) May 10 20:34:58 localhost mfsmaster[20168]: * current
>
* seems that only one copy of the files affected and have lost a disk fail.
* just find a solution to this situation, run mfsfilerepair on each of the
affected files.
* in my case were tens of thousands, so delete the directories where files had
been affected because he was back in the main cluster.
* procedure follows, as the rsyslog an mfscgi and even block the machine by
the number of messages.
1.- /etc/rsyslog.d/mfs
----------- inicio ------------
# => all messages mfs on file:
if ($programname == 'mfsmaster' or $syslogtag == '[mfsmaster]:') then \
-/var/log/mfs/mfs;RSYSLOG_TraditionalFileFormat
if ($programname == 'mfsmaster' or $syslogtag == '[mfsmaster]:') then \
~
--------- fin ----------------------
2.- mkdir -p /var/log/mfs/mfs-old
3.- /etc/init.d/syslog restart
4.- rotate logs every five minutes ?
vi /root/scripts-admin/LogrotateMfs
-----------inicio -----------
/var/log/mfs/mfs {
dateext
ifempty
copytruncate
create 640 root root
olddir /var/log/mfs/mfs-old
sharedscripts
postrotate
/etc/init.d/syslog reload
endscript
lastaction
DATE=`date +%Y%m%d_%H%M` ; \
mv /var/log/mfs/mfs-old/mfs* /var/log/mfs/mfs-old/LOG-MASTER-$DATE ; \
find /var/log/mfs/mfs-old -type f -mtime +1 -exec rm -f {} \;
endscript
}
------------ fin --------------------
5.- crontab -e
*/5 * * * * /usr/sbin/logrotate /root/scripts-admin/LogrotateMfs -f >/dev/null
2>&1
6.- start cluster
7.- mount cluster
7.- start mfscgiserv
list the affected directories and files, take the appropriate decision
according to the quantity and importance, if any backup, etc. ...........
8.- stop mfscgiserv
9.- aply mfssettrashtime 0 to directory and remove.
example.
mfsmount /media/mfs -H 172.26.0.10
cd /media/mfs/
mfssettrashtime -r 0 photos
or mfssettrashtime -r 0 photos/46/45/10
rm -Rf photos or rm -Rf photos/46/45/10
* continue ...........
* users obviously need tools to get listings that apply massively mfscommands
and warning system, disk failures over mail, sms, etc .......
* smartmontools does not work in this case.
* sorry google translator.
|
|
From: Matthias T. <mat...@mt...> - 2010-05-12 17:03:21
|
Moin, I need to export some directories from my Moose cluster via NFS. So I imported the directories via mfsmount on a linux system (2.6.33) and exported it from there via NFS. However the export is not stable. Occasionally requests are terminated with "Stale NFS file handle". After a remount the access is again possible. There is a Userland NFS Server, NFS-Ganesha [1], which was designed for the export of FUSE file systems. NFS-Ganesha wraps the high level functions of fuse.h. My problem is, that mfsmount only uses the low functions from fuse_lowlevel.h and I'm not sure how difficult it is to integrate NFS-Ganesha. What is the best way to export directories from MooseFS via NFS? Has someone used NFS Ganesha with MooseFS? Many thanks Matthias [1] http://nfs-ganesha.sourceforge.net/ |
|
From: Michał B. <mic...@ge...> - 2010-05-11 05:55:14
|
Hi! Probably we know what could have happened. For the beginning please restart the master server (while the chunkservers are still running). If it doesn't help, we'll send you a patch for the master. Kind regards Michał Borychowski From: lwxian_aha [mailto:lwx...@16...] Sent: Tuesday, May 11, 2010 5:34 AM To: moosefs-users Subject: [Moosefs-users] chunkserver can't connect to masterserver today I have new trouble with my new MFS system,chunkserver can't connect to masterserver; my MFS system consist of one masterserver,three chunkserver,every chunkserver with 7.2T diskspace; about 1.4T data and about 7 million files ,every file with 2 copies; MFS version is 1.6.11 OS version CENTOS 5.0 FS is ext3; Following is the error message: [root@localhost mfs]# tail /var/log/messages May 11 10:57:00 localhost mfsmaster[25802]: server 1 (ip: 192.168.10.23, port: 9422): usedspace: 33326616576 (31.04 GiB), totalspace: 140025790464 (130.41 GiB), usage: 23.80% May 11 10:57:00 localhost mfsmaster[25802]: total: usedspace: 33326616576 (31.04 GiB), totalspace: 140025790464 (130.41 GiB), usage: 23.80% May 11 10:57:03 localhost mfsmaster[25802]: CS(192.168.10.21) packet too long (56130653/50000000) May 11 10:57:03 localhost mfsmaster[25802]: chunkserver disconnected - ip: 192.168.10.21, port: 0, usedspace: 0 (0.00 GiB), totalspace: 0 (0.00 GiB) May 11 10:57:08 localhost mfsmaster[25802]: CS(192.168.10.21) packet too long (56130653/50000000) May 11 10:57:08 localhost mfsmaster[25802]: chunkserver disconnected - ip: 192.168.10.21, port: 0, usedspace: 0 (0.00 GiB), totalspace: 0 (0.00 GiB) May 11 10:57:13 localhost mfsmaster[25802]: CS(192.168.10.21) packet too long (56130653/50000000) May 11 10:57:13 localhost mfsmaster[25802]: chunkserver disconnected - ip: 192.168.10.21, port: 0, usedspace: 0 (0.00 GiB), totalspace: 0 (0.00 GiB) May 11 10:57:18 localhost mfsmaster[25802]: CS(192.168.10.21) packet too long (56130653/50000000) May 11 10:57:18 localhost mfsmaster[25802]: chunkserver disconnected - ip: 192.168.10.21, port: 0, usedspace: 0 (0.00 GiB), totalspace: 0 (0.00 GiB) what's happen ?need you help! thanks a lot's 2010-05-05 _____ lwxian_aha |
|
From: lwxian_aha <lwx...@16...> - 2010-05-11 03:34:30
|
today I have new trouble with my new MFS system,chunkserver can't connect to masterserver; my MFS system consist of one masterserver,three chunkserver,every chunkserver with 7.2T diskspace; about 1.4T data and about 7 million files ,every file with 2 copies; MFS version is 1.6.11 OS version CENTOS 5.0 FS is ext3; Following is the error message: [root@localhost mfs]# tail /var/log/messages May 11 10:57:00 localhost mfsmaster[25802]: server 1 (ip: 192.168.10.23, port: 9422): usedspace: 33326616576 (31.04 GiB), totalspace: 140025790464 (130.41 GiB), usage: 23.80% May 11 10:57:00 localhost mfsmaster[25802]: total: usedspace: 33326616576 (31.04 GiB), totalspace: 140025790464 (130.41 GiB), usage: 23.80% May 11 10:57:03 localhost mfsmaster[25802]: CS(192.168.10.21) packet too long (56130653/50000000) May 11 10:57:03 localhost mfsmaster[25802]: chunkserver disconnected - ip: 192.168.10.21, port: 0, usedspace: 0 (0.00 GiB), totalspace: 0 (0.00 GiB) May 11 10:57:08 localhost mfsmaster[25802]: CS(192.168.10.21) packet too long (56130653/50000000) May 11 10:57:08 localhost mfsmaster[25802]: chunkserver disconnected - ip: 192.168.10.21, port: 0, usedspace: 0 (0.00 GiB), totalspace: 0 (0.00 GiB) May 11 10:57:13 localhost mfsmaster[25802]: CS(192.168.10.21) packet too long (56130653/50000000) May 11 10:57:13 localhost mfsmaster[25802]: chunkserver disconnected - ip: 192.168.10.21, port: 0, usedspace: 0 (0.00 GiB), totalspace: 0 (0.00 GiB) May 11 10:57:18 localhost mfsmaster[25802]: CS(192.168.10.21) packet too long (56130653/50000000) May 11 10:57:18 localhost mfsmaster[25802]: chunkserver disconnected - ip: 192.168.10.21, port: 0, usedspace: 0 (0.00 GiB), totalspace: 0 (0.00 GiB) what's happen ?need you help! thanks a lot's 2010-05-05 lwxian_aha |
|
From: lwxian_aha <lwx...@16...> - 2010-05-10 15:17:03
|
today I have new trouble with my new MFS system; my MFS system consist of one masterserver,three chunkserver,every chunkserver with 7.2T diskspace; about 1.4T data and about 7 million files ,every file with 2 copies; today my one chunkserver 's disk is damaged; fist time ,the MFS filesystem is work ok; only problem is lots of file have only on copy; but after about 2 hours , I found that another chunkserver is disappear from the console of mfscgiserv; then, I restart the chunkserver : [root@localhost mfs]# ./sbin/mfschunkserver working directory: /usr/local/mfs/var/mfs lockfile created and locked initializing mfschunkserver modules ... scanning folder /data/mfsdisk1/ ... scanning folder /newdata4/mfs/ ... scanning folder /newdata3/mfs/ ... /data/mfsdisk1/: 277654 chunks found /newdata4/mfs/: 2209988 chunks found /newdata3/mfs/: 2211619 chunks found scanning complete main server module: listen on *:9422 stats file has been loaded mfschunkserver daemon initialized properly it seems the chunkserver is start ok,but it still disappear on the console of mfscgiserv; and then I restart all masterserver and chunkserver ,but after this ,all chunkserver disappear on the console of mfscgiserv; bellow is some message: chunk 1: [root@localhost mfs]# ./sbin/mfschunkserver working directory: /usr/local/mfs/var/mfs lockfile created and locked initializing mfschunkserver modules ... scanning folder /data/mfsdisk1/ ... scanning folder /newdata4/mfs/ ... scanning folder /newdata3/mfs/ ... /data/mfsdisk1/: 277654 chunks found /newdata4/mfs/: 2209988 chunks found /newdata3/mfs/: 2211619 chunks found scanning complete main server module: listen on *:9422 stats file has been loaded mfschunkserver daemon initialized properly [root@localhost mfs]# tail /var/log/messages May 10 20:36:00 localhost mfschunkserver[15720]: connected to Master May 10 20:36:04 localhost mfschunkserver[15720]: testing chunk: /data/mfsdisk1/09/chunk_0000000000287409_00000001.mfs May 10 20:36:10 localhost mfschunkserver[15720]: connecting ... May 10 20:36:10 localhost mfschunkserver[15720]: connected to Master May 10 20:36:14 localhost mfschunkserver[15720]: testing chunk: /newdata4/mfs/D1/chunk_000000000005EAD1_00000001.mfs May 10 20:36:20 localhost mfschunkserver[15720]: connecting ... May 10 20:36:20 localhost mfschunkserver[15720]: connected to Master May 10 20:36:24 localhost mfschunkserver[15720]: testing chunk: /newdata3/mfs/86/chunk_0000000000061086_00000001.mfs May 10 20:36:30 localhost mfschunkserver[15720]: connecting ... May 10 20:36:30 localhost mfschunkserver[15720]: connected to Master chunk 2: [root@localhost mfs]# ./sbin/mfschunkserver working directory: /usr/local/mfs/var/mfs lockfile created and locked initializing mfschunkserver modules ... scanning folder /data/mfsdisk1/ ... scanning folder /newdata2/mfs/ ... scanning folder /newdata1/mfs/ ... /data/mfsdisk1/: 312506 chunks found /newdata2/mfs/: 2182563 chunks found /newdata1/mfs/: 2182481 chunks found scanning complete main server module: listen on *:9422 stats file has been loaded mfschunkserver daemon initialized properly [root@localhost mfs]# tail /var/log/messages May 10 20:33:35 localhost mfschunkserver[25168]: connected to Master May 10 20:33:44 localhost mfschunkserver[25168]: testing chunk: /newdata1/mfs/79/chunk_000000000008A979_00000001.mfs May 10 20:33:45 localhost mfschunkserver[25168]: connecting ... May 10 20:33:45 localhost mfschunkserver[25168]: connected to Master May 10 20:33:54 localhost mfschunkserver[25168]: testing chunk: /data/mfsdisk1/58/chunk_000000000028A658_00000001.mfs May 10 20:33:55 localhost mfschunkserver[25168]: connecting ... May 10 20:33:55 localhost mfschunkserver[25168]: connected to Master May 10 20:34:05 localhost mfschunkserver[25168]: testing chunk: /newdata2/mfs/D6/chunk_00000000000D05D6_00000001.mfs May 10 20:34:05 localhost mfschunkserver[25168]: connecting ... May 10 20:34:05 localhost mfschunkserver[25168]: connected to Master master: [root@localhost mfs]# ./sbin/mfsmaster working directory: /usr/local/mfs/var/mfs lockfile created and locked initializing mfsmaster modules ... loading sessions ... ok sessions file has been loaded exports file has been loaded loading metadata ... loading objects (files,directories,etc.) ... ok loading names ... ok loading deletion timestamps ... ok checking filesystem consistency ... ok loading chunks data ... ok connecting files and chunks ... ok all inodes: 7476300 directory inodes: 19151 file inodes: 7457149 chunks: 7456858 metadata file has been loaded stats file has been loaded master <-> metaloggers module: listen on *:9419 master <-> chunkservers module: listen on *:9420 main master server module: listen on *:9421 mfsmaster daemon initialized properly [root@localhost mfs]# tail /var/log/messages # mass bellow message in the same time May 10 20:34:58 localhost mfsmaster[20168]: * currently unavailable file 158921: photos/47/74/82/r_163042.jpg May 10 20:34:58 localhost mfsmaster[20168]: currently unavailable chunk 00000000004465CA (inode: 4353226 ; index: 0) May 10 20:34:58 localhost mfsmaster[20168]: * currently unavailable file 4353226: photos/46/45/10/857020.jpg May 10 20:34:58 localhost mfsmaster[20168]: currently unavailable chunk 0000000000026B8D (inode: 158922 ; index: 0) May 10 20:34:58 localhost mfsmaster[20168]: * currently unavailable file 158922: photos/47/62/63/m_853671.jpg May 10 20:34:58 localhost mfsmaster[20168]: currently unavailable chunk 00000000004465CB (inode: 4353227 ; index: 0) May 10 20:34:58 localhost mfsmaster[20168]: * currently unavailable file 4353227: photos/46/45/10/s_848116.jpg May 10 20:34:58 localhost mfsmaster[20168]: currently unavailable chunk 0000000000026B8E (inode: 158923 ; index: 0) May 10 20:34:58 localhost mfsmaster[20168]: * currently unavailable file 158923: photos/47/62/63/739789.jpg May 10 20:34:58 localhost mfsmaster[20168]: currently unavailable chunk 00000000004465CC (inode: 4353228 ; index: 0) May 10 20:34:58 localhost mfsmaster[20168]: * currently unavailable file 4353228: photos/46/91/73/936449.jpg May 10 20:34:58 localhost mfsmaster[20168]: currently unavailable chunk 00000000000CBC48 (inode: 158924 ; index: 0) May 10 20:34:58 localhost mfsmaster[20168]: * currently unavailable file 158924: photos/47/74/82/s_171883.jpg May 10 20:34:58 localhost mfsmaster[20168]: currently unavailable chunk 00000000004465CD (inode: 4353229 ; index: 0) May 10 20:34:58 localhost mfsmaster[20168]: * currently unavailable file 4353229: photos/46/45/10/l_848116.jpg May 10 20:34:58 localhost mfsmaster[20168]: currently unavailable chunk 00000000000CBC49 (inode: 158925 ; index: 0) May 10 20:34:58 localhost mfsmaster[20168]: * currently unavailable file 158925: photos/47/74/82/m_156158.jpg May 10 20:34:58 localhost mfsmaster[20168]: currently unavailable chunk 00000000004465CE (inode: 4353230 ; index: 0) May 10 20:34:58 localhost mfsmaster[20168]: * currently unavailable file 4353230: photos/46/91/73/m_928481.jpg May 10 20:34:58 localhost mfsmaster[20168]: currently unavailable chunk 0000000000026B91 (inode: 158926 ; index: 0) May 10 20:34:58 localhost mfsmaster[20168]: * current |
|
From: Michał B. <mic...@ge...> - 2010-05-07 13:43:58
|
The idea is shown on the “About MooseFS” webpage http://www.moosefs.org/about-mfs.html Regarding creating copies during writing process – each of the chunkservers (apart from the last one) connects to the next one and while writing data to its own disk sends the data to the next chunkserver and later sends back the status of the operation. So the extra copies are created at the same moment of writing data. The code is here: mfschunkserver/csserv.c Regarding creating copies when increasing the “goal” parameter or when a copy is lost (e.g. because of disconnecting chunkserver), the master server tells one of chunkservers to send the data to another chunkserver. The code is here: mfschunkserver/replicator.c Kind regards Michał From: 崔慎杰 [mailto:cui...@si...] Sent: Thursday, May 06, 2010 5:13 PM To: moosefs-users Subject: [Moosefs-users] seeking code segment for mfs(about making copies) Hello! I'm a student of China. Now I'm studying distributing file system, my lib use the MFS as a sample to study. I know MFS is a fault tolerant file system, I want to know how MFS build copies on chunkservers.(just as a question of "make a copy just after a write action or wait a free time to make copies.") I have reading the code try to find out where the code about "making copies" but I can't find it out. If you can help me or tell me the localtion of that code segment, I'm really thankful. (My MFS edition is 1.6.11) Thanks! Cui Shenjie 5/6/2010 |
|
From: 崔. <cui...@si...> - 2010-05-06 15:13:15
|
Hello! I'm a student of China. Now I'm studying distributing file system, my lib use the MFS as a sample to study. I know MFS is a fault tolerant file system, I want to know how MFS build copies on chunkservers.(just as a question of "make a copy just after a write action or wait a free time to make copies.") I have reading the code try to find out where the code about "making copies" but I can't find it out. If you can help me or tell me the localtion of that code segment, I'm really thankful. (My MFS edition is 1.6.11) Thanks! Cui Shenjie 5/6/2010 |
|
From: Matthias T. <mat...@mt...> - 2010-05-04 11:03:23
|
Moin, On 20:48 Wed 28 Apr, jose maria wrote: > El Miércoles 28 Abril 2010, Michał Borychowski escribió: > > In order to run MooseFS processes in foreground use '-d' option. It is used > > mainly for debugging but it also starts processes without forking. So we > > assume that it would be somehow possible to use 'daemontools' with this > > option of MooseFS. Yes this works with daemontools but daemontools (svc) uses signals to start, stop and restart daemons. Supervise will restart the daemon if it crashes but I can't use "svc -d /service/mfschunkserver" or something like that. The other tool we can use is monit but it depends on pidfiles. It is all not that important and the mfs daemons are stable enough. > > If you need any further assistance please let us know. We like MooseFS, simplest setup ever and works like a champ. Many thanks Matthias |
|
From: Michał B. <mic...@ge...> - 2010-04-29 08:55:35
|
Hi MarlboroMoo! Thank you for the suggestion. We will think about a solution for quick snapshots removal. Kind regards Michał Borychowski MooseFS Support Manager > -----Original Message----- > From: MarlboroMoo [mailto:mar...@gm...] > Sent: Wednesday, April 28, 2010 12:55 PM > To: moo...@li... > Subject: [Moosefs-users] how to increase speed of remove snapshot ? > > Hi there ! > > i have about 5,000,000 files , and take about 5 second to make > snapshot (via mfsmakesnapshot). > but if i want to remove these snapshot ( rm -rf ), it take a very long time ! > i need to take snapshot every hour , and keep the last three day. > > a directory structure like below: > ========================= > Snapshot-Dir/ > |-- 201004260100/ > |-- 201004260200/ > |-- . > |-- . > |-- 201004280100/ > ========================= > > if i want to remove the snapshot "201004260100" and create new one > "201004280200" > my solution is: > 1) rename the directory "201004260100" to "201004280200" (mv > 201004260100 201004280200) > 2) make a new snapshot to override it. (mfsmakesnapshot > /path/to/source /path/to/201004280200 ) > > maybe we can have a command: mfsrmsnapshot to do this job > cheer ! > -- > Marlboromoo > > ---------------------------------------------------------------------------- -- > _______________________________________________ > moosefs-users mailing list > moo...@li... > https://lists.sourceforge.net/lists/listinfo/moosefs-users |
|
From: jose m. <let...@us...> - 2010-04-28 19:15:22
|
El Miércoles 28 Abril 2010, Michał Borychowski escribió: > Option '-f' has never been used to start MooseFS processes without forking. > > Options '-f' and '-s' have been replaced by parameters 'start', 'restart' > and 'stop'. > > In order to run MooseFS processes in foreground use '-d' option. It is used > mainly for debugging but it also starts processes without forking. So we > assume that it would be somehow possible to use 'daemontools' with this > option of MooseFS. > > But bear in mind that it would work only in chunkservers. If the master > server crashes you need to run "mfsmetarestore" before starting it again. > You could think about writing a script which automatically starts > "mfsmetarestore" and if there are no errors, starts the master server. > > > If you need any further assistance please let us know. > * the problem would be detected when the service crashes or not, and not for a normal stop, start, restart operation. * I'm sorry I'm not a programmer. * I use crontab in all processes and machines, and sysvinit package. * adapt to the particular needs options in runmaster for mfsmaster and mfsmetarestore process. */5 * * * * VigilarMFS.sh RunMaster.sh >/dev/null 2>&1 * options on crontab. or commandline -u user (other user , default root) * commandline option -s (stop process) /usr/local/sbin/VigilarMFS.sh -u user RunMaster.sh /usr/local/sbin/VigilarMFS.sh -s RunMaster.sh * sorry Google Translator ........... |
|
From: MarlboroMoo <mar...@gm...> - 2010-04-28 10:55:09
|
Hi there !
i have about 5,000,000 files , and take about 5 second to make
snapshot (via mfsmakesnapshot).
but if i want to remove these snapshot ( rm -rf ), it take a very long time !
i need to take snapshot every hour , and keep the last three day.
a directory structure like below:
=========================
Snapshot-Dir/
|-- 201004260100/
|-- 201004260200/
|-- .
|-- .
|-- 201004280100/
=========================
if i want to remove the snapshot "201004260100" and create new one
"201004280200"
my solution is:
1) rename the directory "201004260100" to "201004280200" (mv
201004260100 201004280200)
2) make a new snapshot to override it. (mfsmakesnapshot
/path/to/source /path/to/201004280200 )
maybe we can have a command: mfsrmsnapshot to do this job
cheer !
--
Marlboromoo
|
|
From: Michał B. <mic...@ge...> - 2010-04-28 09:44:24
|
Option '-f' has never been used to start MooseFS processes without forking. Options '-f' and '-s' have been replaced by parameters 'start', 'restart' and 'stop'. In order to run MooseFS processes in foreground use '-d' option. It is used mainly for debugging but it also starts processes without forking. So we assume that it would be somehow possible to use 'daemontools' with this option of MooseFS. But bear in mind that it would work only in chunkservers. If the master server crashes you need to run "mfsmetarestore" before starting it again. You could think about writing a script which automatically starts "mfsmetarestore" and if there are no errors, starts the master server. If you need any further assistance please let us know. Kind regards Michał Borychowski MooseFS Support Manager _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Gemius S.A. ul. Wołoska 7, 02-672 Warszawa Budynek MARS, klatka D Tel.: +4822 874-41-00 Fax : +4822 874-41-01 > -----Original Message----- > From: Matthias Teege [mailto:mat...@mt...] > Sent: Tuesday, March 30, 2010 3:07 PM > To: moo...@li... > Subject: [Moosefs-users] supervise mfs daemons > > Moin, > > does anyone use some kind of supervise service for the mfs daemons? We > like to use daemontools but it needs a non forking daemon and the "-f" > flags are deprecated now. Then there is monit which depends on a > pidfile with a pid. The mfs daemons are very stable in our environment > and it is not absolutely essential to supervise the services but we use > daemontools or monit for initial startup. Maybe someone has an idea. > > Many thanks > Matthias > > > ---------------------------------------------------------------------------- -- > Download Intel® Parallel Studio Eval > Try the new software tools for yourself. Speed compiling, find bugs > proactively, and fine-tune applications for parallel performance. > See why Intel Parallel Studio got high marks during beta. > http://p.sf.net/sfu/intel-sw-dev > _______________________________________________ > moosefs-users mailing list > moo...@li... > https://lists.sourceforge.net/lists/listinfo/moosefs-users |
|
From: Michał B. <mic...@ge...> - 2010-04-19 07:06:48
|
The regular commands chown/chgrp should work.
Probalby you mean 999:999 caused by a default option "maproot" in
mfsexports.cfg (see: "man mfsexports.cfg").
If the root is to be the root on the MooseFS (what with the access to the
root on one of the computers and any account on the other computer connected
to the same MooseFS - gives you access to the root on the other compuer) you
should use: "maproot=0:0". If you mount with "nosuid,nodev" and if none of
programs is run under root witih MooseFS, there is no risk.
Kind regards
Michał Borychowski
From: yang xu [mailto:sky...@gm...]
Sent: Monday, April 12, 2010 5:39 PM
To: moo...@li...
Subject: [Moosefs-users] How to change the uid and gid of files stored in
MooseFS?
Well, thank you for creating such a good file system, the MooseFS is
really easy to use, however, I don't know how to set the uid and gid of each
file, the default are 999:999, now in my system, every client shared its
files with others, so how can I set the authority of those files, thank you!
XuYang from China.
|
|
From: Michał B. <mic...@ge...> - 2010-04-14 07:45:52
|
In order to delete files in trash you need to mount on one of the client computers resource "meta" and delete files stored in a "trash" subfolder using a standard command "rm". Data in "trash" are automatically erased after a week so in general you do not have worry about emptying trash. In folders which have lots of files deleted (temps, lockfiles) you can set trashtime to a smaller value, eg. 3600 or even 0. Resource "reserved" should not be cleaned. It contains files which could not be erased because they were still used (opened) by some programs. If you need any further assistance please let us know. Kind regards Michał Borychowski MooseFS Support Manager _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Gemius S.A. ul. Wołoska 7, 02-672 Warszawa Budynek MARS, klatka D Tel.: +4822 874-41-00 Fax : +4822 874-41-01 > -----Original Message----- > From: Ciprian Pinzaru [mailto:cip...@ua...] > Sent: Saturday, April 10, 2010 4:01 PM > To: moo...@li... > Subject: [Moosefs-users] moosefs > > Hi, > I install this program on 65 servers , I have 6TB but my problem is at trash > space. > I don't know how to delete trash space and reserved space. > Thanks, > > -- > ====================================== > = ing. Ciprian Pinzaru = > = = > = UAIC - DCD = > = Phone: +40 (232) 201004 = > = Email: cip...@ua... = > ====================================== |
|
From: Stas O. <sta...@gm...> - 2010-04-13 19:58:33
|
Hi. How MooseFS compares to Hadoop HDFS? Are there any performance benchmarks or reliability tests? Regards. |
|
From: yang xu <sky...@gm...> - 2010-04-12 15:38:58
|
Well, thank you for creating such a good file system, the MooseFS is really easy to use, however, I don't know how to set the uid and gid of each file, the default are 999:999, now in my system, every client shared its files with others, so how can I set the authority of those files, thank you! XuYang from China. |
|
From: Ciprian P. <cip...@ua...> - 2010-04-10 14:19:02
|
Hi, I install this program on 65 servers , I have 6TB but my problem is at trash space. I don't know how to delete trash space and reserved space. Thanks, -- ====================================== = ing. Ciprian Pinzaru = = = = UAIC - DCD = = Phone: +40 (232) 201004 = = Email: cip...@ua... = ====================================== |
|
From: Matthias T. <mat...@mt...> - 2010-03-30 14:20:09
|
Moin, does anyone use some kind of supervise service for the mfs daemons? We like to use daemontools but it needs a non forking daemon and the "-f" flags are deprecated now. Then there is monit which depends on a pidfile with a pid. The mfs daemons are very stable in our environment and it is not absolutely essential to supervise the services but we use daemontools or monit for initial startup. Maybe someone has an idea. Many thanks Matthias |
|
From: Michał B. <mic...@ge...> - 2010-03-23 13:02:49
|
Hi!
The managing (master) server OS has to be POSIX compliant, systems verified
so far are: Linux, FreeBSD, Mac OS X and OpenSolaris. Client computers
(mounting the file system) need to have FUSE installed (also verified on
Linux, FreeBSD, MacOS X and OpenSolaris). So you can choose the system you
are most familiar with.
Yes, MooseFS is 64bit compatible.
If you need any further assistance please let us know.
Kind regards
Michał Borychowski
MooseFS Support Manager
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
Gemius S.A.
ul. Wołoska 7, 02-672 Warszawa
Budynek MARS, klatka D
Tel.: +4822 874-41-00
Fax : +4822 874-41-01
From: wenaideyu wenaideyu [mailto:wen...@gm...]
Sent: Monday, March 22, 2010 1:45 PM
To: moo...@li...
Subject: [Moosefs-users] some questions about moosefs
hello:
i want to buid a moosefs cluster. would you give me some advices about
what operating system should i chose for moosefs which supports moosefs
best? in addtion, can moosefs run in 64bit operating system?
best wishes!
Ai wen
|
|
From: wenaideyu w. <wen...@gm...> - 2010-03-22 12:45:00
|
hello:
i want to buid a moosefs cluster. would you give me some advices about
what operating system should i chose for moosefs which supports moosefs
best? in addtion, can moosefs run in 64bit operating system?
best wishes!
Ai wen
|
|
From: Michał B. <mic...@ge...> - 2010-03-22 09:58:02
|
Thank you for the bug submission. It will be fixed in the next release.
Kind regards
Michał Borychowski
MooseFS Support Manager
From: kuer ku [mailto:ku...@gm...]
Sent: Monday, March 22, 2010 7:35 AM
To: moo...@li...
Subject: [Moosefs-users] bug in mfscommon/main.c
hi, all,
some problem code in mfscommon/main.c ( still in version 1.6.14 ) :
596
597 int check_old_locks(FILE *msgfd,uint8_t runmode,uint32_t timeout) {
598 int lfp;
599 char str[13];
600 uint32_t l;
601 pid_t ptk;
602 char *lockfname;
603
604 lockfname = cfg_getstr("LOCK_FILE",RUN_PATH "/" STR(APPNAME)
".lock");
605 lfp=open(lockfname,O_RDWR);
606 if (lfp<0) {
607 free(lockfname);
^^^^^^^^^^^^^^^^^ WHY free locfname here ???
608 if (errno==ENOENT) { // no old lock file
609 return 0; // ok
610 }
611 syslog(LOG_ERR,"open %s error: %m",lockfname);
^^^^^^^^^^^^^^^^^ WHY refer
lockfname here ????
612 fprintf(msgfd,"open %s error: %s\n",lockfname,errno_to_str());
613 free(lockfname);
614 return -1;
615 }
|
|
From: Michał B. <mic...@ge...> - 2010-03-22 08:04:02
|
From: Marshall Anschutz [mailto:man...@ic...] Sent: Tuesday, March 16, 2010 6:18 PM To: moo...@li... Subject: [Moosefs-users] Mac OS X Snow Leopard mount issues I have successfully built 1.6.13 on Mac OS X Snow Leopard and have started a mfsmaster and mfschunkserver; however, when I try to mount it, I receive an error from mfsmount: mb:test1 tty1$ ./mfsmaster -c mfsmaster.cfg -du mb:test1 tty2$ ./mfschunkserver -c mfschunkserver.cfg -d mb:test1 tty3$ mfsmount testMountPoint/ -o mfsmaster=172.23.10.252,mfsdebug,mfssubfolder=/ mfsmaster register error: Permission denied [MB] Does the master server write anything to the log file? Maybe it is not able to read mfsexports.cfg? For reference, here are the configuration files. I am running as my own user for testing in this test1 directory. mfsmaster.cfg =================== EXPORTS_FILENAME = ./mfsexports.cfg DATA_PATH = ./masterData/ mfschunkserver.cfg: =================== DATA_PATH = ./chunk1/ HDD_CONF_FILENAME = /Users/manschutz/LocalProjects/moosefs/test1/chunk1mfshdd.cfg BIND_HOST = 192.168.110.1 MASTER_HOST = 192.168.110.1 MASTER_PORT = 9420 CSSERV_LISTEN_PORT = 9422 mfsexports.cfg ================== # allow everything but "meta" * / rw,alldirs,maproot=0 # allow "meta" * . rw chunk1mfshdd.cfg ================== /Users/manschutz/LocalProjects/moosefs/test1/chunk1Data1/ /Users/manschutz/LocalProjects/moosefs/test1/chunk1Data2/ [MB] These look OK. If you need any further assistance please let us know. Kind regards Michał Borychowski MooseFS Support Manager _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Gemius S.A. ul. Wołoska 7, 02-672 Warszawa Budynek MARS, klatka D Tel.: +4822 874-41-00 Fax : +4822 874-41-01 Thanks for the assistance! Marshall Anschutz Ice House Productions Senior Software Engineer |
|
From: kuer ku <ku...@gm...> - 2010-03-22 06:34:38
|
hi, all,
some problem code in mfscommon/main.c ( still in version 1.6.14 ) :
596
597 int check_old_locks(FILE *msgfd,uint8_t runmode,uint32_t timeout) {
598 int lfp;
599 char str[13];
600 uint32_t l;
601 pid_t ptk;
602 char *lockfname;
603
604 lockfname = cfg_getstr("LOCK_FILE",RUN_PATH "/" STR(APPNAME)
".lock");
605 lfp=open(lockfname,O_RDWR);
606 if (lfp<0) {
607 free(lockfname);
^^^^^^^^^^^^^^^^^ WHY free locfname here ???
608 if (errno==ENOENT) { // no old lock file
609 return 0; // ok
610 }
611 syslog(LOG_ERR,"open %s error: %m",lockfname);
^^^^^^^^^^^^^^^^^ WHY refer
lockfname here ????
612 fprintf(msgfd,"open %s error: %s\n",lockfname,errno_to_str());
613 free(lockfname);
614 return -1;
615 }
|
|
From: jose m. <aso...@zo...> - 2010-03-16 18:55:08
|
El Martes 16 Marzo 2010, Marshall Anschutz escribió: > mb:test1 tty3$ mfsmount testMountPoint/ -o > mfsmaster=172.23.10.252,mfsdebug,mfssubfolder=/ mfsmaster register error: > Permission denied > > MASTER_HOST = 192.168.110.1 > MASTER_PORT = 9420 * the ip of the machine mfsmaster are different ??? |
84 messages has been excluded from this view by a project administrator.
