RE: errors
"Salyzyn, Mark" <[email protected]>
| Newsgroups | gmane.linux.drivers.aacraid.devel |
|---|---|
| Message-ID | <[email protected]> |
Being that this is a RAID-0, sounds like you are getting a developing drive failure. ext3 is a logging filesystem, and slowly crawls to the offending blocks, rebooting justifies and resets the logging no doubt creating the feeling of stability for a few days. Can you inspect the adapter event logging (excuse my ignorance regarding the tools that report said log please ...)? The killing of the interrupt handler is most likely a separate and serious issue. The modifications for the 2.5.x scsi system changes are an ongoing work in progress. Sincerely -- Mark Salyzyn -----Original Message----- From: Adam Williams [mailto:[email protected]] Sent: Monday, April 07, 2003 9:01 AM To: [email protected] Subject: errors Hello, I have a Dell Poweredge 2600 with a Perc3/Di containing a 100GB raid 0. Frequently, anywhere from 1 to 4 days, the server can't write to disk anymore and prints on TTY1: EXT3-fs error (device sd(8,4)) in ext3_reserve_inode_write: IO Failure I/O error: dev 08:04, sector 0 EXT3-fs error (device sd(8,4)) in ext3_new_inode: IO failure I/O error: dev 08:04 secotor 0 I/O error: dev 08:04 secotor 1050528 EXT3-fs error (device sd(8,4)): ext3_get_inode_loc: unable to read inode block - inode=73040, block=131316 I/O dev 08:04 sector 0 EXT3-fs error (device sd(8,4)) in ext3_reserve_inode_write: IO Failure and on and on, with a lot of sector 0's and other sectors, which numbers seem to be random, as well as the inodes and blocks. But, all of the I/O errors are always from dev 08:04. The server is running the stock 2.4.20 kernel (hand compiled) and redhat 8.0 i386. I can reboot the server and i'll run again fine anywhere from 1-4 days until this error happens again. I upgraded to 2.5.66 as a test, and it gave me a different error message after a couple of days. I don't have that error message written down, but the last line I do remember, which was "AAAAAAIIIIIIEEEEEEEE! killing interrupt handler". With 2.4.20 I can still ssh into the server, but when I do any command, like df, iostat, etc...it says "-bash: /bin/df Input/output error". With 2.4.20 I can hit the rest button, but with 2.5.66 the reset button doesn't work once the box prints the error message on tty1, and I have to inplug the power cables from back. I can't ssh or ping it either. Does anyone have suggestions I can try? Downgrading the kernel to older 2.4 versions still generates the same error. Thanks! Adam _______________________________________________ Linux-aacraid-devel mailing list [email protected] http://lists.us.dell.com/mailman/listinfo/linux-aacraid-devel Please read the FAQ at http://lists.us.dell.com/faq or search the list archives at http://lists.us.dell.com/htdig/ _______________________________________________ Linux-aacraid-devel mailing list [email protected] http://lists.us.dell.com/mailman/listinfo/linux-aacraid-devel Please read the FAQ at http://lists.us.dell.com/faq or search the list archives at http://lists.us.dell.com/htdig/