On Thursday 07 January 2010 17:09:54 Kenneth Gonsalves wrote: > hi, > > I have a client using webfaction. I had set up a django app there 3 years > back. It runs on a particular version of django that I had set up. I have > not had occasion to touch the code (or even ssh to the host) for more than > 6 months now. On occasion my client gets 'nginx bad gateway' error. It > seems to happen fairly often now - perhaps once a month. Since we are only > given an apache instance, the nginx error is obviously on their side. > Usually the site comes up even before a complaint can be filed. 2 days > back this happened. Client complained to me - I asked them to complain to > webfaction. And to my surprise I find the support person says: there is an > error in *your* code. How and nginx bad gateway error appears from *my* > code when I have not touched it for 6 months is not explained. Anyway > before I could ssh in and look at the code, the site came up again. Has > anyone else experienced these problems - first time I have seen such a > casual answer from webfaction! >
I've seen that error when apache isn't running for the client/domain. The only thing I can think of is to monitor the memory usage for the client. If it grows too much then web factions script will shutdown the apache instance. Normally I see this when I'm working in the shell, the site will shutdown (technically all the users process are killed), restarting apache or waiting on the cron script will 'fix' it. Though, your client should be getting emails from web factions script stating the memory usage is too high. Check with them for such emails (or the email the account is registered with at webfaction). The webfaction forums have some memory monitoring scripts and commands. Mike -- Power Company having EMP problems with their reactor
signature.asc
Description: This is a digitally signed message part.
