2008/11/5 Ariel Mauricio Nunez Gomez <[EMAIL PROTECTED]>: > On Tue, Nov 4, 2008 at 7:00 PM, Graham Dumpleton > <[EMAIL PROTECTED]> wrote: >> >> 2008/11/5 Ariel <[EMAIL PROTECTED]>: >> > >> > >> > >> > > I am using mod_wsgi for deployment but recently my apache >> > installation on >> > > ubuntu 8.04 just locks after any start, stop or reload commands. >> > > >> > > I uninstalled it, wiped configuration settings, uninstallled >> > mod_python that >> > > was installed just before it did not respond anymore. >> > >> > It is a known problem that loading mod_python and mod_wsgi at the >> > same >> > time causes problems if Python not installed with a usable shared >> > library. >> > >> > http://code.google.com/p/modwsgi/wiki/InstallationIssues >> > >> > > Is there any other server I can install that works with modwsgi? >> > (I am using >> > > trac and django on the same server configured via mod_wsgi) >> > >> > You mean other than Apache? >> > >> > It seems you haven't explored properly yet why you were having the >> > problems you were. >> > >> > Graham >> > ----- >> > >> > Yeah, I really don't know what to do next, I already disabled >> > mod_python. >> > >> > And tried rebooting my server but even though it says it is going for >> > reboot it does not do so. (It is an Amazon EC2 instance so I am kind >> > of afraid of terminating the instance, although I already made a >> > backup of the most important stuff) >> > >> > Any advice on how to debug this? I checked apache logs but nothing >> > there. >> >> When you say 'rebooting my server', do you mean just Apache or the >> whole system? I don't know how Amazon EC2 works. >> >> If you mean just Apache, do be aware that when removing mod_python, >> generally necessary to do a full 'stop' of Apache. It isn't always >> enough to do an equivalent of 'apachectl restart' because mod_python >> doesn't cleanup properly and tends no to unload correctly on a restart >> where Apache parent process persists. Therefore causes issues for >> mod_wsgi on the restart even though mod_python technically now >> disabled. >> >> If you have a site you can point me to which explains how Amazon EC2 >> is set up, or if you or anyone else can explain what it really is, >> ie., VPS, shared system of some sort, whether you have shell access >> etc, that may help. > > Amazon EC2 is a VPS, but the instance does not have persistent storage, > instead it is always loaded from an image (called AMI), that image is > usually stored in S3. I did one snapshot after I finished configuring my > server, but that was 4 months ago and I am afraid I (may) not have > everything backed up. (Trac repos, mercurial repos, django database > records). > I have full ssh access and root account. > >> >> Anything from the Apache error log files would >> also help, particularly the startup message which lists what modules >> are loaded. >> >> Has anyone else here got experience with Amazon EC2 with or without >> mod_wsgi? >> >> Graham >> > > Well, I did '/etc/init.d/apache2 stop' with no luck, then I just ps -A | > grep apache and killed everything. > After that a '/etc/init.d/apache2 start' just hangs with no error on apache > logs. > > I check system messages and it turns out: > [EMAIL PROTECTED]:/var/log/apache2# tail -f /var/log/messages > Nov 4 23:59:47 pinax kernel: 4gb seg fixup, process bash (pid 23795), cs:ip > 73:0016b817 > Nov 4 23:59:48 pinax kernel: printk: 2 messages suppressed. > Nov 4 23:59:48 pinax kernel: 4gb seg fixup, process sshd (pid 23767), cs:ip > 73:0047cc6c > Nov 4 23:59:54 pinax kernel: printk: 109 messages suppressed. > > Maybe it is directly related to Amazon machines: > http://groups.google.com/group/ec2ubuntu/browse_thread/thread/a145b228d83c0b71
Hmmm, maybe it is an Amazon issue. Help me to understand the problem a bit better though. When you do run: /etc/init.d/apache2 start are you saying that that command itself hangs, or that any request against Apache hangs, or the whole system hangs? How many Apache processes do you see running? How much memory are they using? Do the Apache processes appear to be doing straight away and are then replaced with new ones? Ie., a fork bomb due to startup failure? How have you configured mod_wsgi, are you trying to use daemon mode? Graham --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "modwsgi" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/modwsgi?hl=en -~----------~----~----~----~------~----~------~--~---
