Zope2.0.0 final - hung.
I have a weird one here. I've been using Zope 2.0.0 final, binary release, Solaris 5.6 since it came out. It's been fine. Now it's hung. 'top' shows (my username only) load averages: 0.03, 0.07, 0.11 13:45:59 297 processes: 296 sleeping, 1 on cpu CPU states: 97.1% idle, 0.0% user, 0.8% kernel, 2.0% iowait, 0.0% swap Memory: 512M real, 8712K free, 739M swap in use, 813M swap free PID USERNAME THR PRI NICE SIZE RES STATE TIME CPU COMMAND 6508 nnle 1 18 0 2024K 1488K cpu/2 0:00 0.27% top 5586 nnle 1 23 0 2072K 1848K sleep 0:00 0.01% tcsh 2377 nnle 7 33 0 11M 8696K sleep 2:27 0.00% python 2390 nnle 1 33 0 2680K 1872K sleep 0:02 0.00% mysql 2376 nnle 4 16 0 4296K 1200K sleep 0:00 0.00% python 6203 nnle 1 23 0 2064K 1816K sleep 0:00 0.00% tcsh 17 nnle 1 23 0 1960K 1328K sleep 0:00 0.00% tcsh 6434 nnle 1 34 0 2664K 2192K sleep 0:00 0.00% lynx ie Python isn't taking up large amounts of CPU. I've tried connecting to the server using Lynx etc. and no joy. The FTP server on port 8021 is also not responding. I haven't *as yet* killed the processes and started new ones, as I'd *really* like to know what it is I've done. The only thing I've done is to use the binary release of python to do some MySQL work. I haven't installed Andy Dustmans' MySQLdb-0.1.0 yet. I used the same python installation that the Zope server is using. Is it possible I've done something to the 'interpreter lock' or somesuch... If the outcome is that I should leave the Zope python binary well alone whilst Zope is running - then that is completely icebox with me. I hate to say this, but .... help!. tone. ------ Dr Tony McDonald, FMCC, Networked Learning Environments Project http://nle.ncl.ac.uk/ The Medical School, Newcastle University Tel: +44 191 222 5888 Fingerprint: 3450 876D FA41 B926 D3DD F8C3 F2D0 C3B9 8B38 18A2
In article <v04205510b40697054ad2@[128.240.198.13]>, Tony McDonald <tony.mcdonald@ncl.ac.uk> writes
I have a weird one here. I've been using Zope 2.0.0 final, binary release, Solaris 5.6 since it came out. It's been fine.
Now it's hung. 'top' shows (my username only) load averages: 0.03, 0.07, 0.11 13:45:59 297 processes: 296 sleeping, 1 on cpu CPU states: 97.1% idle, 0.0% user, 0.8% kernel, 2.0% iowait, 0.0% swap Memory: 512M real, 8712K free, 739M swap in use, 813M swap free
PID USERNAME THR PRI NICE SIZE RES STATE TIME CPU COMMAND 6508 nnle 1 18 0 2024K 1488K cpu/2 0:00 0.27% top 5586 nnle 1 23 0 2072K 1848K sleep 0:00 0.01% tcsh 2377 nnle 7 33 0 11M 8696K sleep 2:27 0.00% python 2390 nnle 1 33 0 2680K 1872K sleep 0:02 0.00% mysql 2376 nnle 4 16 0 4296K 1200K sleep 0:00 0.00% python 6203 nnle 1 23 0 2064K 1816K sleep 0:00 0.00% tcsh 17 nnle 1 23 0 1960K 1328K sleep 0:00 0.00% tcsh 6434 nnle 1 34 0 2664K 2192K sleep 0:00 0.00% lynx
ie Python isn't taking up large amounts of CPU. I've tried connecting to the server using Lynx etc. and no joy. The FTP server on port 8021 is also not responding.
I haven't *as yet* killed the processes and started new ones, as I'd *really* like to know what it is I've done.
The only thing I've done is to use the binary release of python to do some MySQL work. I haven't installed Andy Dustmans' MySQLdb-0.1.0 yet. I used the same python installation that the Zope server is using. Is it possible I've done something to the 'interpreter lock' or somesuch...
Are both pythons hung? Were you interfering with a db that the Zope has open or somesuch? Have you tried the monitor_client.py inside ZServer/medusa? I run lots of things beside my win32 Zope and haven't locked up Zope2+CVS as yet. Running other python processes shouldn't do anything nasty. I assume that the binary python is 1.5.2 my RH6.0 came with 1.5.1, but I replaced it.
If the outcome is that I should leave the Zope python binary well alone whilst Zope is running - then that is completely icebox with me.
I hate to say this, but .... help!. tone. ------ Dr Tony McDonald, FMCC, Networked Learning Environments Project http://nle.ncl.ac.uk/ The Medical School, Newcastle University Tel: +44 191 222 5888 Fingerprint: 3450 876D FA41 B926 D3DD F8C3 F2D0 C3B9 8B38 18A2
_______________________________________________ Zope maillist - Zope@zope.org http://www.zope.org/mailman/listinfo/zope
(To receive general Zope announcements, see: http://www.zope.org/mailman/listinfo/zope-announce
For developer-specific issues, zope-dev@zope.org - http://www.zope.org/mailman/listinfo/zope-dev )
-- Robin Becker
At 3:36 pm +0100 16/9/99, Robin Becker wrote:
Are both pythons hung? Were you interfering with a db that the Zope has open or somesuch? Have you tried the monitor_client.py inside ZServer/medusa? I run lots of things beside my win32 Zope and haven't locked up Zope2+CVS as yet. Running other python processes shouldn't do anything nasty. I assume that the binary python is 1.5.2 my RH6.0 came with 1.5.1, but I replaced it.
Thanks for theReply Robin, Here's more of the story; I couldn't figure out how to test whether 'both' pythons were hung. I found that I could log into the medusa FTP server easily, but once I did a 'dir' nothing came back. I killed the processes and restarted Zope (./start) again (top shows two python processes running). It's possible that I was tinkering with a database that Zope was using, and (as I was doing MySQL stuff with my Python scripts) I presume I was using the same python MySQL library. I'll have a look at Andy Dustmans' new patch to ZMySQLDA and see if I can figure around that. I didn't try the monitor_client.py voodoo (I don't understand how to use it correctly). This is the only time that Zope 2 has locked up on my Solaris box, on my LinuxPPC box at home, it's happened a few times (this is with Solaris 2.0.0 binary release, Python 1.5.2 and 2.0.0 source release, Python 1.5.2 respectively). cheers, tone. ------ Dr Tony McDonald, FMCC, Networked Learning Environments Project http://nle.ncl.ac.uk/ The Medical School, Newcastle University Tel: +44 191 222 5888 Fingerprint: 3450 876D FA41 B926 D3DD F8C3 F2D0 C3B9 8B38 18A2
Tony McDonald wrote:
I have a weird one here. I've been using Zope 2.0.0 final, binary release, Solaris 5.6 since it came out. It's been fine.
I haven't *as yet* killed the processes and started new ones, as I'd *really* like to know what it is I've done.
The only thing I've done is to use the binary release of python to do some MySQL work. I haven't installed Andy Dustmans' MySQLdb-0.1.0 yet. I used the same python installation that the Zope server is using. Is it possible I've done something to the 'interpreter lock' or somesuch...
This could be the problem if you are using other code in the same process space as Zope. If it's a different process, then they are unrelated.
If the outcome is that I should leave the Zope python binary well alone whilst Zope is running - then that is completely icebox with me.
I hate to say this, but .... help!.
I would restart and see if you can recreate it. -Michel
At 12:30 pm -0400 16/9/99, Michel Pelletier wrote:
The only thing I've done is to use the binary release of python to do some MySQL work. I haven't installed Andy Dustmans' MySQLdb-0.1.0 yet. I used the same python installation that the Zope server is using. Is it possible I've done something to the 'interpreter lock' or somesuch...
This could be the problem if you are using other code in the same process space as Zope. If it's a different process, then they are unrelated.
Is there any way I can find this out? My knowledge of Python innards is pretty slim.
If the outcome is that I should leave the Zope python binary well alone whilst Zope is running - then that is completely icebox with me.
I hate to say this, but .... help!.
I would restart and see if you can recreate it.
I've restarted it, and nothing has gone wrong yet. If it does, I'll let the list know. cheers Tone ------ Dr Tony McDonald, FMCC, Networked Learning Environments Project http://nle.ncl.ac.uk/ The Medical School, Newcastle University Tel: +44 191 222 5888 Fingerprint: 3450 876D FA41 B926 D3DD F8C3 F2D0 C3B9 8B38 18A2
participants (3)
-
Michel Pelletier -
Robin Becker -
Tony McDonald