Hi folks,

 

I have one of the original Unraid systems built from the specs Tom gave us including the hot swappable drive trays.  I am currently on 4.3.3 and I have the parity drive and 8 data drives in my system.  Yesterday I noticed my system was no longer available on the network so I went to the tower and the lights on the drive trays were on solid for drives 6 and 7, but no other activity.  So I powered down the tower and brought it back up.  Everything came up fine and the system did a parity check overnight finishing with no errors.  Earlier today I noticed the shares were not available on the network again, so I checked out the tower again with the same symptoms.  Drives 6 and 7 lights on solid and no other activity.  I powered down again, but now when I power up drives 6, 7, and 8 come up red and labeled as missing in the GUI.

 

Any ideas where to start troubleshooting this one?

 

Thanks,

 

Doug

Hi folks,

 

I have one of the original Unraid systems built from the specs Tom gave us including the hot swappable drive trays.  I am currently on 4.3.3 and I have the parity drive and 8 data drives in my system.  Yesterday I noticed my system was no longer available on the network so I went to the tower and the lights on the drive trays were on solid for drives 6 and 7, but no other activity.  So I powered down the tower and brought it back up.  Everything came up fine and the system did a parity check overnight finishing with no errors.  Earlier today I noticed the shares were not available on the network again, so I checked out the tower again with the same symptoms.  Drives 6 and 7 lights on solid and no other activity.  I powered down again, but now when I power up drives 6, 7, and 8 come up red and labeled as missing in the GUI.

 

Any ideas where to start troubleshooting this one?

 

Thanks,

 

Doug

Post a copy of your syslog.  Instructions in the wiki.  It may have clues.

 

Then, I'd probably re-seat the disk controller common to disks 6,7, and 8 (if they are on a single controller) but be sure to do that with the server powered down and unplugged from the wall outlet (or at least switched off)

Thanks for the quick reply Joe.  I have the server powered down right now, so I'll try reseating the controller first and after I boot up I'll post the syslog.

 

Thanks!

 

Doug

Any syslog is better than none, but if possible, the syslog after you see those lights in error would be best.

Drive 5,6,7, and 8 are missing.  I don't remember now if 5 was originally missing or if my problem is getting worse...

 

Here is the syslog:

 

 

I just verified with my son.  Disk 5 was okay earlier and is now missing on this boot.  This is after reseating both of the added PCI Controllers.

System reports all IDE, 4 onboard ports (hda thru hdd), and 2 Promise cards with 4 ports on each.  The drives attached to the 4 onboard connectors are fine, but each Promise card reported one of the following lines:

kernel: ide3: Wait for ready failed before probe !

kernel: ide4: Wait for ready failed before probe !

 

Since both Promise cards are behaving with the same type of problem, it is probably not the Promise cards at fault (unless it is bad power being supplied).

 

Device inventory:
Aug 30 00:28:38 Tower emhttp: pci-0000:00:1f.1-ide-0:0 (hda) ata-WDC_WD3200JB-00KFA0_WD-WCAMR4023898
Aug 30 00:28:38 Tower emhttp: pci-0000:00:1f.1-ide-0:1 (hdb) ata-WDC_WD2000JB-00GVA0_WD-WCALL1125745
Aug 30 00:28:38 Tower emhttp: pci-0000:00:1f.1-ide-1:0 (hdc) ata-WDC_WD2500BB-22GUA0_WD-WCAL72248242
Aug 30 00:28:38 Tower emhttp: pci-0000:00:1f.1-ide-1:1 (hdd) ata-MAXTOR_STM3320620A_9QF7N3GT
Aug 30 00:28:38 Tower emhttp: pci-0000:02:00.0-ide-0:0 (hde) ata-WDC_WD2500JB-22GVC0_WD-WCAL75281427
Aug 30 00:28:38 Tower emhttp: restart_md_driver: stat pci-0000:02:00.0-ide-0:1: No such file or directory
Aug 30 00:28:38 Tower emhttp: restart_md_driver: stat pci-0000:02:00.0-ide-1:0: No such file or directory
Aug 30 00:28:38 Tower emhttp: restart_md_driver: stat pci-0000:02:00.0-ide-1:1: No such file or directory
Aug 30 00:28:38 Tower emhttp: restart_md_driver: stat pci-0000:02:01.0-ide-0:0: No such file or directory

 

This table is formed based on the drives found, and what your disk.cfg and super.dat indicate should be found.  The last 4 of your drives were not found.  Based on their slot numbering, it looks like you had 4 drives on one Promise card, of which only hde was found, but not hdf or hdg or hdh.  It looks like you had one more drive on the other Promise card, hdi, also not found.

 

It looks like a problem that is common to those 4 drives, perhaps a backplane containing just them, or a common power cable, etc...  It could also be a power supply problem, perhaps it is failing, one rail has maybe failed.  The drives themselves are probably fine, although it is possible that one or more were damaged at the time this failure was occurring.  Hopefully not.

Do you guys know of any hardware monitoring tools that could boot off of a flash drive to help determine where the problem might be?  Something that can monitor the rails on the Power supply, check the ide controllers, etc?

 

Thanks,

 

Doug

My server has two power supplies and all four of those drives are on the second power supply.  So it looks like that must be the problem.  A new Corsair 750 to replace both is on the way and I'm crossing my fingers that there was no data corruption on multiple drives.

 

Thanks for the help,

 

Doug

Got the new power supply today and the server is working beautifully again. 

 

Thanks for the help guys.

 

Doug