Unsolved
This post is more than 5 years old
5 Practitioner
•
274.2K Posts
0
2680
SQL DDBEA 3.0 trans log, delayed start
I have a customer whose SQL DDBEA 3.0 trans log backups take up to 50 minutes, while they take only 5 to 10 writing a direct SQL dump to disk.
Our BEA agent seems to spend at least 30 minutes inspecting every DB on the host, one at time, before it begins writing a trans log backup. Is this normal behavior? Is there anything we can do to skip these checks? The trans log backup is taking longer than the DB stripes this way.
37790:(pid 5856):Database model backup level forced to full
115705:(pid 5856):Only full backup is supported on msdb database.
115872:(pid 5856):Backup level for database msdb is forced to full.
.
87746:(pid 5856):Recovery Model of Database Journyx is simple. Backup of database Journyx will be skipped.
87746:(pid 5856):Recovery Model of Database JOM_Ent_EDM is simple. Backup of database JOM_Ent_EDM will be skipped.
87746:(pid 5856):Recovery Model of Database Benlysta_EDM is simple. Backup of database Benlysta_EDM will be skipped.
87746:(pid 5856):Recovery Model of Database Allergan_EDM is simple. Backup of database Allergan_EDM will be skipped.
87746:(pid 5856):Recovery Model of Database WebAnalysis is simple. Backup of database WebAnalysis will be skipped.
87746:(pid 5856):Recovery Model of Database Sanofi_Ent_EDM is simple. Backup of database Sanofi_Ent_EDM will be skipped.
87746:(pid 5856):Recovery Model of Database Amgen_EDM is simple. Backup of database Amgen_EDM will be skipped.
87746:(pid 5856):Recovery Model of Database Regeneron_EDM is simple. Backup of database Regeneron_EDM will be skipped.
87746:(pid 5856):Recovery Model of Database Dendreon_EDM is simple. Backup of database Dendreon_EDM will be skipped.
87746:(pid 5856):Recovery Model of Database BMS_PAF_EDM is simple. Backup of database BMS_PAF_EDM will be skipped.
87746:(pid 5856):Recovery Model of Database Genzyme_ENT_EDM is simple. Backup of database Genzyme_ENT_EDM will be skipped.
ble1
14.3K Posts
0
September 21st, 2016 13:00
From your log times can't be seen. Since NW module for SQL and SQL DBBEA share alot, I assume from log above you already use variable to skip DBs in simple mode when running TLOG backup. I use that as well (but with real NW module) and this is fast. The only thing which may cause delay (though not so big) is check for health of transaction log sequence (eg. if some other backup broke it so that you don't have broken chain when doing restore). I heard that in release 9 (which might be equivalent to what you use) this can be switched off, but still I never seen this before so I doubt this is your issue.
If I were you, I would enable some debugging to see what it does exactly and would ship those logs to support (unless it would be clear from them where the issue is).