| Title: | DIGITAL UNIX (FORMERLY KNOWN AS DEC OSF/1) |
| Notice: | Welcome to the Digital UNIX Conference |
| Moderator: | SMURF::DENHAM |
| Created: | Thu Mar 16 1995 |
| Last Modified: | Fri Jun 06 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 10068 |
| Total number of notes: | 35879 |
I have a customer running TruCluster on their AS8200/AS4100 machine with V4.0B. This is a new configuration and the TruCluster has been installed for only a week. Recently the got their system panic at every reboot and they were not able to use the system. The panic is "alloc_mcell: bad mcell free list". It seemed that the mcell free list of a certain Advfs was corrupted. We than asked the customer to boot to single user mode and suppress the startup of TruCluster. Rebooting to multiuser is now sucessful. We thus suspect the problem come from the shared SCSI disk on the HSZ40. Later the customer confirmed us that the advfs was able to be mounted but touching any file will panic the system. Is there anyone seen similar case? I've cross posted this on ADVFS_SUPPORT note conference. Thanks, Mickwid.
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 9940.1 | one crash like this reported in CANASTA | HAN::HALLE | Volker Halle MCS @HAO DTN 863-5216 | Mon May 26 1997 02:26 | 28 |
Mickwid,
according to the CANASTA world-wide case database, there has been a
crash like this on node KITCHE (reported by AutoCLUE for Digital UNIX).
Eric Schott @ZKO should be able to answer what has happened on that
node...
SERVER_0001350 OGO UNIDENTIFIED ACLUE-KITCHE.ZK 23-APR-1997 09:33:54.00 ~
SERVER_0001350 OGO V4.0B alloc_mcell: bad mcell free list
SERVER_0001350 OGO boot panic advfs_sad alloc_mcell bmt_alloc_mcell
stg_alloc_new_mcell get_first_mcell alloc_append_stg
alloc_stg add_stg stg_add_stg_no_cow
stg_add_stg rbf_add_stg fs_create_file msfs_mkdir mkdir
syscall _Xsyscall
SERVER_0001350 OGO alloc_mcell:3038, 0xfffffc00002fb070] nil
In the future, please always obtain the crash-data file from ANY crash
and send it to the CANASTA Mail Server for analysis and documentation
of the crash footprint. The mails erver will automatically send you a
list of crashes with the same footprints, if there is no rule match
in CANASTA.
Please see note 8919 for infomration on how to use the CANASTA Mail
Server. Information on how to access to world-wide CANASTA crash
footprint database can be found in note SPECXN::CANASTA 239.
Volker.
| |||||