June 7, 2013 at 12:12 pm
So I have been working on a large scale replication project that has suddenly come to a standstill and I’m hoping someone here has some advice. My problem is that adding new subscriptions and publications take 15+ minutes and about 30 of the published databases are now behind like 1+ day.
Environment: (All SQL 08 R2 Latest SP)
Published DB’s and Distribution database:
2 node cluster (24 proc server 256GB ram on a high-end SAN)
1,214 Published DBs
Subscriber Server
Same as above
Can Transactional Replication handle this? Right now the CPU and ram is fine and I don’t see an IO load but everything is very very slow! I’m not even sure what to look for. I do notice the only large table in the distribution is the commands table and I can’t even re-index it because it’s always locked. The distribution table has file groups across different LUNs. Should I be creating multiple distribution databases? Should I have the distribution database on its own physical server? I’m at a loss.
Thanks!
June 13, 2013 at 3:47 pm
Hi,
How many subscribers have you got.. Are the push or pull ? 1200+ publishers seems a lot for 1 distribution db. Have you Checked the waits, locking, blocking etc ? I would hazard a guess that there's some contention from the log readers & distribution agents.
Cheers
Vultar
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply