Messages in this thread | | | Date | Sat, 12 Apr 1997 21:05:47 -0400 (EDT) | From | Jon Lewis <> | Subject | Re: ext2 filesystem corruption?!?!?? |
| |
On Fri, 4 Apr 1997, Jeff Garzik wrote:
> Here's my machine config on which I've been having the filesystem > corruption problems.
I'm also seeing the ext2 corruption occasionally, primarily on just 2 systems. The first is our news server:
P90, 128MB RAM, 2 NCR 810 (using BSD ported driver, tagged queuing off, no_atime in use, 5mhz sync), 3c509. It has: Host: scsi0 Channel: 00 Id: 00 Lun: 00 Vendor: CONNER Model: CP30540 545MB3.5 Rev: AEB8 Type: Direct-Access ANSI SCSI revision: 02 Host: scsi0 Channel: 00 Id: 02 Lun: 00 Vendor: IBM Model: DPES-31080 !t Rev: S31R Type: Direct-Access ANSI SCSI revision: 02 Host: scsi0 Channel: 00 Id: 04 Lun: 00 Vendor: IBM Model: DPES-31080 !t Rev: S31K Type: Direct-Access ANSI SCSI revision: 02 Host: scsi1 Channel: 00 Id: 01 Lun: 00 Vendor: MICROP Model: 3243-19MZ Q4D Rev: HT02 Type: Direct-Access ANSI SCSI revision: 02 Host: scsi1 Channel: 00 Id: 03 Lun: 00 Vendor: DEC Model: DSP5300S Rev: 427L Type: Direct-Access ANSI SCSI revision: 02
All the above are internal on granite digital cables (it's not a bad cable problem :). It has 6 feeds and stays pretty busy when not filled up.
The other is our tape backup host, P100, 48mb RAM, 1 NCR 810 (BSD driver, tagged queuing off), SMC Ultra. Host: scsi0 Channel: 00 Id: 00 Lun: 00 Vendor: IBM Model: DALS-3540 !s Rev: S60E Type: Direct-Access ANSI SCSI revision: 02 Host: scsi0 Channel: 00 Id: 01 Lun: 00 Vendor: IBM Model: DORS-32160 !# Rev: WA3E Type: Direct-Access ANSI SCSI revision: 02 Host: scsi0 Channel: 00 Id: 04 Lun: 00 Vendor: PIONEER Model: CD-ROM DRM-602X Rev: 2902 Type: CD-ROM ANSI SCSI revision: 02 [luns 1-5 on the Pioneer clipped] Host: scsi0 Channel: 00 Id: 06 Lun: 00 Vendor: ARCHIVE Model: Python 28388-XXX Rev: 5.AC Type: Sequential-Access ANSI SCSI revision: 02
All devices are internal except the CD.
I've seen ext2 corruption with a variety of 2.0.x kernels. The tape host was running 2.0.29, and today when I got to work, I found a screen full of:
EXT2-fs error (device 08:16): ext2_new_block: Free blocks count corrupted for block group 48 EXT2-fs error (device 08:16): ext2_new_block: Free blocks count corrupted for block group 48 EXT2-fs error (device 08:16): ext2_new_block: Free blocks count corrupted for block group 48 EXT2-fs error (device 08:16): ext2_new_block: Free blocks count corrupted for block group 48
I've upgraded it to 2.0.30 now with tagged queuing on and 10mhz sync, plus Gerard's ext2 debugging patch...sort of hoping it will happen again. The partition where the corruption was is /home (on the DORS), which is where the tape scripts log stuff during network backups. That logging is about all the disk activity this system sees at night.
------------------------------------------------------------------ Jon Lewis <jlewis@fdt.net> | Unsolicited commercial e-mail will Network Administrator | be proof-read for $199/hr. ________Finger jlewis@inorganic5.fdt.net for PGP public key_______
| |