[Spacewalk-list] Spacewalk 1.6 Database problem

[Sascha Bendix]

Hi,

I just recognized the same problem last week because my spacewalk was getting slowly slower and slower. At that point in time it was a spacewalk 1.6 with PostgreSQL on a CentOS 6.2 system

I first tried to figure out which processes are responsible for this behavior. After I found this thread I did an upgrade to 1.7 and CentOS 6.3. The problem is definitely less.

But after the weekend I see again 20 processes with "postgres: spaceuser spaceschema [local] UPDATE waiting". They are blocked by 16 "postgres: spaceuser spaceschema [local] idle in transaction" processes that hold locks. The blocked queries all look like this:

update rhnServerAction set status = 1,  pickup_time = current_timestamp, remaining_tries = 3 - 1 where action_id = 34826 and server_id = 1000010711

In general I don't think it's a problem with the performance of the database. The queries just seem to stay and pile up.

I currently don't know how to get deeper into the problem. Can somebody imagine a way to get the process ids that are responsible for these sessions? 

Does somebody has an another idea?

Regards,

Sascha

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.