Unsolved
This post is more than 5 years old
2 Posts
0
12214
December 29th, 2003 13:00
Fatal IO errors on Perc 3/Di
We have several 2650's running RHAS 3 and 2.1. After a few days of heavy I/O, they all start failing with these errors on the console:
EXT3-fs error (device sd(8,3)) in ext3_reserver_inode_write: IO failure
I/O error: dev 08:03, sector 0
I/O error: dev 08:03, sector 8388640
EXT3-fs error (device sd(8,3)): ext3_get_inode_loc: unable to read inode block
inode=524372, block=1048580
There are no flashing lights indicating a disk problem and the systems are non-responsive and hung at this point.
Once the systems are rebooted, they are fine for another few days.
I am running:
Component Revisions
-------------------
CLI: 2.7-1 (Build #4944)
API: 2.7-1 (Build #4944)
Miniport Driver: 2.7-1 (Build #3170)
Controller Software: 2.7-1 (Build #3170)
Controller BIOS: 2.7-1 (Build #3170)
Controller Firmware: (Build #3170)
Red Hat/Adaptec aacraid driver (1.1.2 Oct 3 2003 18:09:55) -- RHAS 3.0 servers
It seems like there are dozens of people on http://lists.us.dell.com/pipermail/linux-poweredge with the similar problems, but there is no resolution. Any ideas?
EXT3-fs error (device sd(8,3)) in ext3_reserver_inode_write: IO failure
I/O error: dev 08:03, sector 0
I/O error: dev 08:03, sector 8388640
EXT3-fs error (device sd(8,3)): ext3_get_inode_loc: unable to read inode block
inode=524372, block=1048580
There are no flashing lights indicating a disk problem and the systems are non-responsive and hung at this point.
Once the systems are rebooted, they are fine for another few days.
I am running:
Component Revisions
-------------------
CLI: 2.7-1 (Build #4944)
API: 2.7-1 (Build #4944)
Miniport Driver: 2.7-1 (Build #3170)
Controller Software: 2.7-1 (Build #3170)
Controller BIOS: 2.7-1 (Build #3170)
Controller Firmware: (Build #3170)
Red Hat/Adaptec aacraid driver (1.1.2 Oct 3 2003 18:09:55) -- RHAS 3.0 servers
It seems like there are dozens of people on http://lists.us.dell.com/pipermail/linux-poweredge with the similar problems, but there is no resolution. Any ideas?
No Events found!
sc-hick
2 Posts
0
March 19th, 2004 00:00
I have the EXACT same issue on a 6300 running RH9.0. Using the Perc2 4 channel (aacraid). I've performed low level format, rebuilt containers, upgraded all bios and firmware on motherboard, raidcard, and drives....running out of ideas.
Have you made any progress on finding a resolution? I'm having to pwr cycle server once a day....is getting real old real quick.
Jean-Philippe H
1 Message
0
March 31st, 2004 17:00
I have the exact same problem with a PowerEdge 2650 and RH AS 2.1. I found a lot of similar problem, but no solutions. I ran a diagnostic from dell and the only thing it find is an error with the Power supply senser, which I guess, is the blue/red light that you can plug in the back, but not sure I have to check.
I would appreciate any help on that too!
Thanks.
sc-hick
2 Posts
0
March 31st, 2004 17:00
My final resolve was to swap the raid controller for a Megaraid controller. The problem has not surfaced since. Thanks,
xanol
2 Posts
0
March 31st, 2004 19:00
The Dell lists and Redhat bugzilla point to the cause of the issue as being either in controller firmware or accraid driver. Dell and Redhat support were of little help resolving this for us and mostly pointed fingers at each other regarding who should fix it. The fact is though, when those two variables change (ie. disable 3/Di and install 3/Dc) the problem goes away.