[Search for users]
[Overall Top Noters]
[List of all Conferences]
[Download this site]
| Title: | LSM |
|
| Moderator: | SMURF::SHIDERLY |
|
| Created: | Mon Jan 17 1994 |
| Last Modified: | Fri Jun 06 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 803 |
| Total number of notes: | 2852 |
A customer is running 3.2D-1 & LSM. On several occasions vold has gotten
into an 'inaccessible' state for periods up to 15 minutes. This occurs when
an LSM disk disappears, either through device failure or removal from a
shelf. Basically, vold will be inaccessible while some other command is
attempting to query the nonexistant disk. Is there anyway to reduce the
timeout period for such queries?
Most recently, the customer discovered a bug in ASE 1.3 where the asecdb
wasn't
being updated to remove LSM disks from a configuration. So he'd remove a disk
from LSM and modify the service to remove it. Then when he used voldiskadm
to modify another portion of his LSM config, it would hang vold as it tried
to do a 'voldisk list' on a disk that really wasn't in LSM anymore. This
has been addressed by an ASE workaround, but the customer would like a way
of making LSM commands to nonexistant disks timeout faster. Ideas/suggestions
appreciated.
TIA
-Jason
[Posted by WWW Notes gateway]
| T.R | Title | User | Personal
Name | Date | Lines
|
|---|