EVMS problem
|
View:
New views
3 Messages
—
Rating Filter:
Alert me
|
 |
EVMS problem
by hardhead
::
Rate this Message:
I have a RAID5 managed by EVMS using LVM2 running on a Linux server. Earlier today the system went down unexpectedly and when the box came up one of the disks had a bad block so the raid is running in degraded mode. Unfortunately, when I try to add a spare disk to the array--I get a failure message: Oct 06 23:16:02 jumbo _0_ Engine: engine_user_message: Message is: Engine: Plug-in MDRaid5RegMgr returned error 19 when committing changes for region md/md1 during phase 3. I began to research this error message--and I *think* the problem is the MD object isn't activating. Without that activation, I can not manage the RAID (so I can't add a spare). When running a simple activate command in the CLI, the following appears in my evms-engine.log file: _5_ Engine: is_object_change_pending: Change pending: Object lvm2/media/home needs to be activated. _5_ Engine: is_object_change_pending: Change pending: Object lvm2/media/media needs to be activated. _5_ Engine: is_object_change_pending: Change pending: Object md/md1 needs to be activated. _5_ Engine: is_volume_change_pending: Change pending: Volume /dev/evms/home needs to be activated. _5_ Engine: is_volume_change_pending: Change pending: Volume /dev/evms/media needs to be activated. _3_ Engine: engine_ioctl_object: ioctl to object md/md1 failed with error code 5: Input/output error _2_ MDRaid5RegMgr: md_ioctl_run_array: md_ioctl_run_array: IOCTL failed, region(md/md1) rc(5) _3_ Engine: activate_object: Error code 5 activating object md/md1: Input/output error _3_ Engine: engine_ioctl_object: ioctl to object md/md1 failed with error code 16: Device or resource busy In /var/log/messages I see lots of: md: array md1 already has disks! When I run mdadm --detail /dev/md1 I get: /dev/md1: Version : 00.90.03 Creation Time : Mon May 15 20:55:42 2006 Raid Level : raid5 Device Size : 244195840 (232.88 GiB 250.06 GB) Raid Devices : 7 Total Devices : 6 Preferred Minor : 1 Persistence : Superblock is persistent Update Time : Sat Oct 6 22:26:37 2007 State : active, degraded Active Devices : 6 Working Devices : 6 Failed Devices : 0 Spare Devices : 0 Layout : left-asymmetric Chunk Size : 256K UUID : 66774afe:e3d47cc5:57d24caa:8eb4f32b Events : 0.30322321 Number Major Minor RaidDevice State 0 253 0 0 active sync /dev/evms/.nodes/hdc1 1 0 0 - removed 2 253 1 2 active sync /dev/evms/.nodes/sdb1 3 253 2 3 active sync /dev/evms/.nodes/sdc1 4 253 3 4 active sync /dev/evms/.nodes/sdd1 5 253 4 5 active sync /dev/evms/.nodes/sde1 6 253 5 6 active sync /dev/evms/.nodes/sdf1 I have tried mdadm --stop /dev/md1 before running evms (or evmsn) and I keep running into the same problem. Any help on this issue would be much appreciated. I know I *could* potentially rebuild the raid with mdadm -a /dev/sda1, but I'm not sure how EVMS would react to that, and I'm not sure my two volumes stored on the raid (/dev/evms/media and /dev/evms/home) would remain intact. Also--when I run ls /dev/evms/.nodes I see: hdc hdc1 sda sdb sdb1 sdc sdc1 sdd sdd1 sde sde1 sdf sdf1 I notice sda1 doesn't yet exist, so I can't manually use mdadm to add sda1. Please let me know if any more information would be helpful--and if there is any way to solve this problem. Thanks, -Michael Beam ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Evms-devel mailing list Evms-devel@... To subscribe/unsubscribe, please visit: https://lists.sourceforge.net/lists/listinfo/evms-devel |
|
|
Re: EVMS problem
by Steinar H. Gunderson
::
Rate this Message:
On Sat, Oct 06, 2007 at 11:35:25PM -0400, hardhead wrote:
> I know I *could* potentially rebuild the raid with mdadm -a > /dev/sda1, but I'm not sure how EVMS would react to that FWIW, if you use /dev/evms/.nodes/sda1 instead, it's perfectly safe. I recently saw your problem, but didn't find a good solution. /* Steinar */ -- Homepage: http://www.sesse.net/ ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Evms-devel mailing list Evms-devel@... To subscribe/unsubscribe, please visit: https://lists.sourceforge.net/lists/listinfo/evms-devel |
|
|
Re: EVMS problem
by hardhead
::
Rate this Message:
I was able to fix the problem with a few steps with help from folks at the gentoo forums. The discussion is detailed here:
http://forums.gentoo.org/viewtopic-t-595202.html Ultimately, I found some discussion from the IRC logs that led to the commands that helped: mdadm --stop /dev/md1 mdadm -A /dev/md1 -f /dev/evms/.nodes/hdc1 /dev/evms/.nodes/sdb1 /dev/evms/.nodes/sdc1 /dev/evms/.nodes/sdd1 /dev/evms/.nodes/sde1 /dev/evms/.nodes/sdf1 Cheers, -Michael On 10/8/07, Steinar H. Gunderson <sgunderson@...
> wrote: On Sat, Oct 06, 2007 at 11:35:25PM -0400, hardhead wrote: ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Evms-devel mailing list Evms-devel@... To subscribe/unsubscribe, please visit: https://lists.sourceforge.net/lists/listinfo/evms-devel |
| Free Forum Powered by Nabble | Forum Help |
