Plesk 11.5 - 502 bad gateway

[ManuelG2k]

Hello,
I have Plesk Panel 11.5.30 Update #3 on Linux ubuntu Server 12.0.4.2 LTS

When I try to export a database from phpmyadmin, I receive this error:

502 bad Gateway

I find this error on /var/log/sw-cp-server/error_log

2013/07/16 09:31:20 [error] 4298#0: *33397 recv() failed (104: Connection reset by peer) while reading response header from upstream, client: 88.149.137.195, server: , request: "POST /domains/databases/phpMyAdmin/export.php HTTP/1.1", upstream: "fastcgi://unix:/var/run/sw-engine.sock:", host: "mydomain.com:8443", referrer: "https://mydomain.com:8443/domains/databases/phpMyAdmin/db_export.php?db=la-raia_it&server=1&token=df81cd4565060e1a281ce8c774b371e6"


Can you help me ?

 

[MisterU]

Do you have any related records in syslog? What is the size of you db?

 

[ManuelG2k]

Do you have any related records in syslog? What is the size of you db?

Database size is 21 Mb
No error on syslog
I reboot the server and now works :|

 

[Silvester van der Leer]

Database size is 21 Mb
No error on syslog
I reboot the server and now works :|

Can you repeat the export? on multiple servers i cannot export databases all times (first time it works mostly, other times not)
Same problem : bad gateway.

De panel.log is giving strage errors:

[17-Jul-2013 10:57:36 Europe/Berlin] PHP Fatal error: Cannot redeclare (null)() (previously declared in /opt/psa/admin/htdocs/domains/databases/phpMyAdmin/libraries/export/sql.php:9) in /opt/psa/admin/htdocs/domains/databases/phpMyAdmin/libraries/export/sql.php on line 325

On line 9 of the sql.php:
if (! defined('PHPMYADMIN')) {
exit;
}

 

[ManuelG2k]

Can you repeat the export? on multiple servers i cannot export databases all times (first time it works mostly, other times not)
Same problem : bad gateway.

De panel.log is giving strage errors:

[17-Jul-2013 10:57:36 Europe/Berlin] PHP Fatal error: Cannot redeclare (null)() (previously declared in /opt/psa/admin/htdocs/domains/databases/phpMyAdmin/libraries/export/sql.php:9) in /opt/psa/admin/htdocs/domains/databases/phpMyAdmin/libraries/export/sql.php on line 325

On line 9 of the sql.php:
if (! defined('PHPMYADMIN')) {
exit;
}

Same error.

[17-Jul-2013 10:16:26 Europe/Berlin] PHP Fatal error: Cannot redeclare (null)() (previously declared in /opt/psa/admin/htdocs/domains/databases/phpMyAdmin/libraries/export/sql.php:9) in /opt/psa/admin/htdocs/domains/databases/phpMyAdmin/libraries/export/sql.php on line 325

after restarting the server everything works

 

[Silvester van der Leer]

after restarting the server everything works

Reboot or just apache/psa/nginx?
And for how long does it work ?

Bit strange to me, seems more like a repeative bug to me

 

[ManuelG2k]

Reboot or just apache/psa/nginx?
And for how long does it work ?

Bit strange to me, seems more like a repeative bug to me

I try to restart apache, psa and ngix, but not work.
After reboot entire server, works.

 

[Silvester van der Leer]

I try to restart apache, psa and ngix, but not work.
After reboot entire server, works.

Ok, but i don't want to reboot the whole server for such a bug, a restart of apache/ plesk /nginx should be sufficient.
Can you try to export mysql databases a few times? and to restart plesk / apache / nginx in between.

I keep getting the error even after psa stopall / start

 

[Silvester van der Leer]

Ok, but i don't want to reboot the whole server for such a bug, a restart of apache/ plesk /nginx should be sufficient.

Tried a reboot on a testserver, it worked for 1 time, after that, same errors.

 

[Nikolay.]

Basically, if anything like reboot helps in such situation, restart of sw-engine service should help as well. Please check. Also post your /usr/local/psa/admin/conf/php.ini.

 

[Silvester van der Leer]

Basically, if anything like reboot helps in such situation, restart of sw-engine service should help as well. Please check. Also post your /usr/local/psa/admin/conf/php.ini.

Yes, but i tried the reboot now on the testserver, same problem occurs

short_open_tag = On
y2k_compliance = Off
output_buffering = Off
max_execution_time = 600
max_input_time = 600
memory_limit = 128M
max_file_uploads = 99999
max_input_vars = 2000

error_reporting = E_ALL & ~E_NOTICE & ~E_STRICT & ~E_DEPRECATED
log_errors = On
display_errors = Off
warn_plus_overloading = Off
expose_php = Off

variables_order = "ECGPS"
register_argc_argv = On
post_max_size = 2147483647
magic_quotes_gpc = Off
magic_quotes_runtime = Off

include_path = "/opt/psa/admin/plib:/opt/psa/admin/externals"
upload_tmp_dir = "/tmp"
upload_max_filesize = 2147483647

apc.stat = 1
apc.shm_size = 100M
apc.max_file_size = 500K
apc.slam_defense = 0
apc.enabled = 1

swkey.repository_dir = "/etc/sw/keys"

psasem.semfile = "/opt/psa/var/psasem.sem"

zend_extension = "/usr/lib/php/modules/sw-engine/ioncube_loader_lin_5.4.so"

Almost every time i'm getting the 520 bad gateway when i export a database (even a empty one) in phpmyadmin

 

[ManuelG2k]

for me, after restarting the server, the export from phpmyadmin works

 

[Silvester van der Leer]

for me, after restarting the server, the export from phpmyadmin works

Here sometimes after reboot, not stable enough, even if i can get a dump 1 time, it doesn't mean i can do a second time.

 

[Nikolay.]

In /usr/local/psa/admin/conf/php.ini try setting apc.enabled = 0 and then doing # service sw-engine restart. If this helps, then revert your changes and instead set up apc.filters to exclude phpMyAdmin code.

@topic starter: if you're allowed to, please rename the thread to something more appropriate. I spent quite some time searching for it to post this message.

 

[Dree]

I'm having the same problem.
Restarting doesn't help, disabling apc doesn't work either.
Anyone else have any suggestions?

 

[Silvester van der Leer]

I'm having the same problem.
Restarting doesn't help, disabling apc doesn't work either.
Anyone else have any suggestions?

I've "fixed" it for now with setting
keepalive_timeout 1800;

in /etc/nginx/nginx.conf

but that should not be needed.

 

[Dree]

Thanks for your answer.
This doesn't fix it for me, and I'm also pretty sure Plesk doesn't use this config file, cause it uses sw-cp-server

 

[Silvester van der Leer]

Thanks for your answer.
This doesn't fix it for me, and I'm also pretty sure Plesk doesn't use this config file, cause it uses sw-cp-server

You have to restart nginx ofcourse, but strangely it affects thats phpmyadmin is working again.

I think more people should have the same issue if they are going to check it (try dumping a medium size database)

 

[Nikolay.]

Altering /etc/nginx/nginx.conf will have no effect on phpMyAdmin (provided you use one bundled with Plesk) since it works on sw-cp-server/sw-engine. Just as Dree have said.

 

[Silvester van der Leer]

Altering /etc/nginx/nginx.conf will have no effect on phpMyAdmin (provided you use one bundled with Plesk) since it works on sw-cp-server/sw-engine. Just as Dree have said.

Will revert the change back then, and try again this afternoon.