A forum for technical support discussion related to Fogbugz.
FogBugz Technical Support
FogBugz Version 7.0.32 (DB 716, Build 1020)
Windows 2003 std , database platform is SQL 2000 - on a seperate server.
the software has only recently started experiencing this problem, - all hosts are fully up to date etc -
this issue existed prior to upgrading to the latest version.
A problem was reported in that the email notifications were no longer being sent from our Fogbugz installation and that there were occasional web server errors being thrown such as "SQL server does not exist" or access denied.
have narrowed the problem down to something that heartbeat.asp is doing - or not doing .
for now - we have left the FogBugz Maintenance Service disabled - such that the website at least remains stable
If there are rows in the mailqueue and we manually invoke heartbeat.asp - it detects a lock and using the break lock functionality, everything is then fine - it sends the emails and returns an http response of "+ Broke the heartbeat lock"
The next time heartbeat.asp is called - either via a browser or the maintenance service, there is very quickly a 'system wide' issue that eventually leads to heartbeat.asp and also the website occasionally itself timing out , returning an error - "SQL server does not exist" or access denied.
Have been through a lot of this forum and have eliminated all the likely suspects - there are no firewall or permissions issues etc .
on the db server - everything fine - typically a couple of connections open from fogbugz - other apps are able to access the server ok.
On the web server - prior to running heartbeat.asp, netstat-a indicates no problems.
After running heartbeat, netstat shows about 4000 tcp connections from the web server to the sql server in a TIME_WAIT state - so this is the reason why the error "SQL server does not exist" or access denied gets returned.
It suggests to me that heartbeat is in some kind of loop, and failing to communicate with the database properly - or receiving some kind of unexpected response?
has anyone come across a problem like this in fogbugz - or - any suggestions as to how to proceed?
no pop3 mailboxes
Index Progress 99%
Total Indexed 31966
Total Unindexed 35
Thursday, October 8, 2009
not that familiar with your schema - best guess as per the query below shows that there are a few bugs with a large number of bugevents
select top 10 ixBug,count(*) from bugevent group by ixBug order by 2 desc
Take a look at those bugs, and it's likely there's some mail loop or runaway process making them that's jamming up the works.
You'll probably have to clean out the repeat BugEvent rows...
Michael H. Pryor
Thursday, October 8, 2009
This topic is archived. No further replies will be accepted.Other recent topics