PVE Slow because oh the Hard Disk?

[nick]

Hi All,

I have a problem with a Small server (Xeon 3220 with 4GBRAM) that have Hard Disks connected to SATA - without RAID or something.

Until last week everything was OK. On this server was installed a Zimbra Server (Qemu with CentOS5.2 allocates 2GB RAM) and a Windows XP Pro SP3 with 1024 RAM allocated where it's running a TrendMicro OfficeScan Server.

Now, I move Zimbra to another server from cluster and I leave only WinXP...and all the time I receive error - on screen and log:

Nov 15 06:25:42 pve_master kernel: sd 2:0:0:0: [sda] 312581808 512-byte hardware sectors (160042 MB)
Nov 15 06:25:42 pve_master kernel: sd 2:0:0:0: [sda] Write Protect is off
Nov 15 06:25:42 pve_master kernel: sd 2:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Nov 15 06:25:42 pve_master kernel: sd 2:0:0:0: [sda] 312581808 512-byte hardware sectors (160042 MB)
Nov 15 06:25:42 pve_master kernel: sd 2:0:0:0: [sda] Write Protect is off
Nov 15 06:25:42 pve_master kernel: sd 2:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Nov 15 06:25:51 pve_master kernel:         res 51/40:00:a4:68:da/40:00:10:00:00/e0 Emask 0x9 (media error)
Nov 15 06:25:51 pve_master kernel: ata3.00: configured for UDMA/133
Nov 15 06:25:51 pve_master kernel: ata3: EH complete
Nov 15 06:25:54 pve_master kernel:         res 51/40:00:a4:68:da/40:00:10:00:00/e0 Emask 0x9 (media error)
Nov 15 06:25:55 pve_master kernel: ata3.00: configured for UDMA/133
Nov 15 06:25:55 pve_master kernel: ata3: EH complete
Nov 15 06:25:58 pve_master kernel:         res 51/40:00:a4:68:da/40:00:10:00:00/e0 Emask 0x9 (media error)
Nov 15 06:25:58 pve_master kernel: ata3.00: configured for UDMA/133
Nov 15 06:25:58 pve_master kernel: ata3: EH complete
Nov 15 06:26:01 pve_master kernel:         res 51/40:00:a4:68:da/40:00:10:00:00/e0 Emask 0x9 (media error)
Nov 15 06:26:01 pve_master kernel: ata3.00: configured for UDMA/133
Nov 15 06:26:01 pve_master kernel: ata3: EH complete
Nov 15 06:26:04 pve_master kernel:         res 51/40:00:a4:68:da/40:00:10:00:00/e0 Emask 0x9 (media error)
Nov 15 06:26:04 pve_master kernel: ata3.00: configured for UDMA/133

and

sd 2:0:0:0: [sda] 312581808 512-byte hardware sectors (160042 MB)
Nov 15 06:34:45 pve_master kernel: sd 2:0:0:0: [sda] Write Protect is off
Nov 15 06:34:45 pve_master kernel: sd 2:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Nov 15 06:34:45 pve_master kernel: sd 2:0:0:0: [sda] 312581808 512-byte hardware sectors (160042 MB)
Nov 15 06:34:45 pve_master kernel: sd 2:0:0:0: [sda] Write Protect is off
Nov 15 06:34:45 pve_master kernel: sd 2:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Nov 15 06:35:00 pve_master kernel:         res 51/40:00:a4:68:da/40:00:10:00:00/e0 Emask 0x9 (media error)
Nov 15 06:35:00 pve_master kernel: ata3.00: configured for UDMA/133
Nov 15 06:35:00 pve_master kernel: ata3: EH complete
Nov 15 06:35:03 pve_master kernel:         res 51/40:00:a4:68:da/40:00:10:00:00/e0 Emask 0x9 (media error)
Nov 15 06:35:03 pve_master kernel: ata3.00: configured for UDMA/133
Nov 15 06:35:03 pve_master kernel: ata3: EH complete
Nov 15 06:35:06 pve_master kernel:         res 51/40:00:a4:68:da/40:00:10:00:00/e0 Emask 0x9 (media error)
Nov 15 06:35:06 pve_master kernel: ata3.00: configured for UDMA/133
Nov 15 06:35:06 pve_master kernel: ata3: EH complete
Nov 15 06:35:09 pve_master kernel:         res 51/40:00:a4:68:da/40:00:10:00:00/e0 Emask 0x9 (media error)
Nov 15 06:35:10 pve_master kernel: ata3.00: configured for UDMA/133
Nov 15 06:35:10 pve_master kernel: ata3: EH complete
Nov 15 06:35:13 pve_master kernel:         res 51/40:00:a4:68:da/40:00:10:00:00/e0 Emask 0x9 (media error)
Nov 15 06:35:13 pve_master kernel: ata3.00: configured for UDMA/133
Nov 15 06:35:13 pve_master kernel: ata3: EH complete
Nov 15 06:35:16 pve_master kernel:         res 51/40:00:a4:68:da/40:00:10:00:00/e0 Emask 0x9 (media error)
Nov 15 06:35:16 pve_master kernel: ata3.00: configured for UDMA/133
Nov 15 06:35:16 pve_master kernel: sd 2:0:0:0: [sda] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK
Nov 15 06:35:16 pve_master kernel: sd 2:0:0:0: [sda] Sense Key : Medium Error [current] [descriptor]
Nov 15 06:35:16 pve_master kernel: Descriptor sense data with sense descriptors (in hex):
Nov 15 06:35:16 pve_master kernel:        72 03 11 04 00 00 00 0c 00 0a 80 00 00 00 00 00
Nov 15 06:35:16 pve_master kernel:        10 da 68 a4
Nov 15 06:35:16 pve_master kernel: sd 2:0:0:0: [sda] Add. Sense: Unrecovered read error - auto reallocate failed

The system respond Slow, and the I/O is more then 40%. How I can trace the problem?

 

[udo]

Hi,
buy fast a new disk!

I think media error say all.

Udo

 

[nick]

I try to move all machines from this server to another and for I machine I receive error (see attach file).

How I can transfer this last machine?

 

[udo]

Hi Nick,
you can try to copy the disk "offline" via a rescue-boot-cd. dd_rescue tried to copy the disk raw to another (blank disk). But with media error your qcow-file will be damage. Perhaps you can use something, but definitivly not all - in a qcow-file are only data - no wasted space, so an error must tend data!
I think it's time to restore from backup (hopefully you have one).

Udo

 

[nick]

I have a backup...I'm not worry about this! I want to learn from this issue until I replace disk and restore the machine!

I'm not sure if is only a disk issue or it's also a qcow2 issue. Few weeks ago I have another disk failure to another PVE server and I manual recover all my work.

Tomorrow I will try the same procedure...

 

[dietmar]

I have a backup...I'm not worry about this! I want to learn from this issue until I replace disk and restore the machine!

I know this does not help now, but you should really use a RAID setup to protect your data - it willl make your life easier ;-)