Hi there,
we also have loop break couple months ago and it stays mystery.
F740 single head, 5.1.2, FC-AL 18 GB disks in 6 shelves.
We test all disk shelves one shelf at time with 2 disks and all works ok, but adding all 6 shelves loop stays broken. Swapping one shelf nro3 and problem disappears.
What disk fw do you have at that time, I have pretty old.
-TomiR- tomi.rautio@nokia.com
-----Original Message----- From: EXT bentele@mtu-friedrichshafen.com [mailto:bentele@mtu-friedrichshafen.com] Sent: 10. November 1999 11:18 To: toasters@mathworks.com; jay@cimedia.com Subject: AW: 'Loop break detected' followed by failover Importance: High
Hi Jay,
we also had a ' isp2100_timeout]: Resetting ISP2100 in slot ..' a few weeks ago. F740 , DOT 5.2.3, FC-AL 18 GB, Gbit Card.
!! Be careful !!
Replacing motherboard and FC-Adapter-Card at last solved the problem, after we replaced all cables, all shelves of this loop, and about 3 disks.
Mit freundlichen Grüßen
Markus Bentele MTU Friedrichshafen GmbH
- 07541-90-2654
- bentele@mtu-friedrichshafen.com
Von: Jay Soffian[SMTP:jay@cimedia.com] Gesendet: Mittwoch, 10. November 1999 08:07 An: toasters@mathworks.com Betreff: 'Loop break detected' followed by failover
I've opened a case with NOW on this.
One of my filers (F740/512MB/28x9GB disks/DOT 5.2.3)
tonight just did
this:
Tue Nov 9 19:00:01 EST [subzero: statd]: 7:00pm up 44 days, 13:45 2570782327 NFS ops, 0 CIFS ops, 0 HTTP ops Tue Nov 9 19:16:46 EST [subzero: isp2100_main]: Loop break
detected on
ISP2100 in slot 1. Tue Nov 9 19:16:50 EST [subzero: isp2100_timeout]:
Resetting ISP2100 in
slot 1 Tue Nov 9 19:19:14 EST [subzero/viking: cf_takeover]:
relog syslog Tue
Nov 9 19:17:15 EST [subzero: isp2100_timeout]: Resetting
ISP2100 in slot
1 Tue Nov 9 19:19:14 EST [subzero/viking: cf_takeover]:
relog syslog Tue
Nov 9 19:17:27 EST [subzero: isp2100_timeout]: Resetting
ISP2100 in slot
1 Tue Nov 9 19:19:15 EST [subzero/viking: asup_main]:
Cluster Notification
mail sent
No core dump, I found the filer at an 'ok' prompt.
I rebooted the filer, did a cf giveback a few minutes later and all appears fine now.
Anything recommended besides checking that all cables are tight and all drives are fully seated?
I couldn't find any bugs relating to this on NOW (except a
bug in the
Diagnostics not resetting the FC-AL loop properly when a loop break occurs). I've heard rumors that some F740's were shipped with bad on-board FC-AL controllers, but this F740 has been in operation for over a year w/o any problems (it's only been clustered since June though).
Suggestions?
j.