hi all,
as said above, our w2003 guest which is the AD server too, has some strange behaviour - as far as i can tell - since when we moved it from an old pve 1.7 to the new 2.1
i'm not sure at all that is pve related, more i'm almost sure that's not pve related, but i've read of other 2003 problems here and there so who knows maybe someone here had also this issue...
the issue is: despite the fact that everything works normally on this server, our backup system (hp openview data protector 5.5) is no more able to backup the "configuration" settting of this server.
the data protector "configuration" setting is a "virtual" setting for backing up as a whole all the vital info about the system and the AD itself (including registry, dns, etc). the backup worked flawlessly until the server was on pve 1.7, but since we moved on pve 2.1 we have this problem nearly every night (backup runs at 1:00) and the backuplog reports
"
List of Objects That Did Not Complete Successfully Object Type Client Mountpoint Description Status # Errors # Warnings Device ____________________________________________________________________________________________________________________________________ WinFS server /CONFIGURATION CONFIGURATION: Failed 0 1 disco_f
[Major] From: BSM@amleto "backup" Time: 21/09/2012 0.07.11 [61:1002] The VBDA named "CONFIGURATION:" on host server reached its inactivity timeout of 7200 seconds. The agent on host will be shutdown. "
i've other windows guests on the pve 2.1 server (one 2000 and one 2008r2) and occasionally also the w2000 guest shows the same behaviour for the backup (but a regular folder backup, not "configuration") of one _partition_ of the three available on the same virtual (iscsi/lvm/raw/ide) disk, never the other partitions on the same disk. but this issue raises only sometimes while the w2003 issue is permanent now... w2003 disk is iscsi/lvm/raw/virtio.
i tried to understand why that backup job lasts for two hours (the timout limit) and then fails, with data protector debug methods (not that easy), to my best, but all i can understand is that the local disk agent can't talk to the disk until fails... so i'm asking if anybody here has hints. i'm also thinking to restore a 1.7 pve server and backup restore there the machine there just to see if this bug can be pve version (with its kvm, etc) related... can you suggest any other test/approach to solve this problem?
what can it be? thanks to all.
Marco
as said above, our w2003 guest which is the AD server too, has some strange behaviour - as far as i can tell - since when we moved it from an old pve 1.7 to the new 2.1
i'm not sure at all that is pve related, more i'm almost sure that's not pve related, but i've read of other 2003 problems here and there so who knows maybe someone here had also this issue...
the issue is: despite the fact that everything works normally on this server, our backup system (hp openview data protector 5.5) is no more able to backup the "configuration" settting of this server.
the data protector "configuration" setting is a "virtual" setting for backing up as a whole all the vital info about the system and the AD itself (including registry, dns, etc). the backup worked flawlessly until the server was on pve 1.7, but since we moved on pve 2.1 we have this problem nearly every night (backup runs at 1:00) and the backuplog reports
"
List of Objects That Did Not Complete Successfully Object Type Client Mountpoint Description Status # Errors # Warnings Device ____________________________________________________________________________________________________________________________________ WinFS server /CONFIGURATION CONFIGURATION: Failed 0 1 disco_f
[Major] From: BSM@amleto "backup" Time: 21/09/2012 0.07.11 [61:1002] The VBDA named "CONFIGURATION:" on host server reached its inactivity timeout of 7200 seconds. The agent on host will be shutdown. "
i've other windows guests on the pve 2.1 server (one 2000 and one 2008r2) and occasionally also the w2000 guest shows the same behaviour for the backup (but a regular folder backup, not "configuration") of one _partition_ of the three available on the same virtual (iscsi/lvm/raw/ide) disk, never the other partitions on the same disk. but this issue raises only sometimes while the w2003 issue is permanent now... w2003 disk is iscsi/lvm/raw/virtio.
i tried to understand why that backup job lasts for two hours (the timout limit) and then fails, with data protector debug methods (not that easy), to my best, but all i can understand is that the local disk agent can't talk to the disk until fails... so i'm asking if anybody here has hints. i'm also thinking to restore a 1.7 pve server and backup restore there the machine there just to see if this bug can be pve version (with its kvm, etc) related... can you suggest any other test/approach to solve this problem?
what can it be? thanks to all.
Marco
Last edited: