Hi all, I have 44 supermicro PDSMi servers running linux 2.6.17.13. Each of these systems have a Promise Super Trak EX16350 controller with 16 WDC WD5000ABYS-01TNA disks connected. The controllers have BIOS 2.9.022 installed and the stex driver used is version 2.9.0.16. These servers are used in a backup-to-disk environment and are connected via NFS to 4 backup servers. Now every few days, mostly at night, one of these servers hang for about 40 minutes which can be defined as not accepting any TCP/IP connections; no crontab jobs (sar) are run. After this period that can vary from 5 minutes to 60 minutes spontaneously the system becomes available again and disconnects the stex controller device not the disks. This also occurs with the latest version of Linux 2.6.23.1 but in this version it has as a more negative effect as also both devices created on these disks are offlined and the system hangs for the same amount of time. With the newer version of the kernel I also get an handshake failure message. As the root filesystem is on these disks too I cannot login into the server and have to reboot the system after what the system comes up like nothing has happened. With the newer version of the kernel and stex driver the system is also unstable when I allow the system to start in SMP then I cannot even create a ext3 filesystem as I get a kernel panic in add_list telling me the pointers are corrupt. Question has anybody else encountered these problem? In what direction do I have to seek is this a issue with the stex driver or stex firmware? Thanks in advance Mike. - To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
