Our backup software, Time Navigator from Atempo, uses SMS to backup NSS volumes. Backup is often unsucessful, volumes become invisible from SMS randomly and we need to restart manualy smdrd (rcnovell-smdrd restart), then our software can see volumes again. In this case, the backup process stops and return an error. There is no way to continue it, we need to restart backup from the begining.

The problem is on each of our OES2 SP3 server (physical or VMware guest).

We use OES2 SP3 with the latest patches, and TSAFS parameters are the following :
- Read buffer size : 131072 bytes
- Read threads per job : 8
- Read threads allocation : 100 %
- Read ahead throttle : 2
- Cache memory threshold : 10
- Caching and clustering are disabled
- TSA mode = Netware, according to 3rd party backup software

We tried to change the set of these values, unsuccessfully.

If you have an idea, thanks for your help.