plesk 10 in vz on centos

[jmackenz]

Hello,

I set up a new container in vz. Chose the .plesk set up profile. I then ran the parallels auto installer. wget -O - http://autoinstall.plesk.com/one-click-installer | sh



Now when I go to the control panel login page for plesk, after logging in I get :

the server is temporarily unable to service your request due to maintenance downtime or capacity problems please try again later

any ideas on where to look>?

http error log reads :

ls Panel' does NOT match server name!?
[Mon Mar 28 00:43:43 2011] [warn] RSA server certificate CommonName (CN) `Parall els Panel' does NOT match server name!?
[Mon Mar 28 00:43:43 2011] [warn] Init: SSL server IP/port conflict: default-69_ xxx_xxx_xxx:443 (/usr/local/psa/admin/conf/generated/13012581080.11424100_server .include:101) vs. horde.webmail:443 (/usr/local/psa/admin/conf/generated/xxxxxxx8 1080.11424100_horde.include:48)
[Mon Mar 28 00:43:43 2011] [warn] Init: You should not use name-based virtual hosts in conjunction with SSL!!

[jmackenz]

[root@hosting logs]# pwd
/usr/local/psa/admin/logs
[root@hosting logs]# ls
httpsd_access_log httpsd_access_log.processed panel.log sitebuilder.log


there is no httpsd_error_log like I see referred to elsewhere, and panel.log is empty.

[jmackenz]

[root@hosting logs]# uname -a
Linux hosting 2.6.18-028stab077.1 #1 SMP Mon Nov 1 19:25:11 MSK 2010 i686 i686 i386 GNU/Linux

[paulie]

Hi jmackenz,

As of Plesk 9 (and 10) the Plesk Control Panel is provided by lightppd which logs to /var/log/sw-cp-server/error_log (or something pretty close to that!).

Paste the output from there to here and/or try "/etc/init.d/psa restart", but its more than likely a bigger problem than that.

Also please can you paste the output of "cat /proc/user_beancounters".

[jmackenz]

/etc/init.d/psa restart
cat /var/log/sw-cp-server/error_log

Code: Select all

2011-04-12 14:00:39: (log.c.135) server stopped
2011-04-12 14:00:39: (log.c.75) server started

Code: Select all

Version: 2.5
       uid  resource           held    maxheld    barrier      limit    failcnt
  5216931:  kmemsize        3001210    3001210   11055923   11377049          0
            lockedpages           0          0        512        512          0
            privvmpages       90388      90391     965536     969632         69
            shmpages            955        955      21504      21504          0
            dummy                 0          0 2147483647 2147483647          0
            numproc              51         51        240        240          0
            physpages         35698      35702          0 2147483647          0
            vmguarpages           0          0      33792 2147483647          0
            oomguarpages      35698      35702 2147483647 2147483647          0
            numtcpsock           24         24        360        360          0
            numflock             11         11        188        206          0
            numpty                1          1         16         16          0
            numsiginfo            0          1        256        256          0
            tcpsndbuf        215040     217280    1720320    2703360          0
            tcprcvbuf        214208     214208    1720320    2703360          0
            othersockbuf      17920      17920    1048576    2097152          0
            dgramrcvbuf           0          0     262144     262144          0
            numothersock         20         21        360        360          0
            dcachesize            0          0    3409920    3624960          0
            numfile            1836       1837       9312       9312          0
            dummy                 0          0          0          0          0
            dummy                 0          0          0          0          0
            dummy                 0          0          0          0          0
            numiptent            10         10        128        128          0

[faris]

Hmmm.. I'd give it more memory in case that's the problem.

Faris.

[jmackenz]

Sorry for the delay on this but I rolled an excavator down a hill , so the system has 4gb physical RAM , should this be enough for Virtuozzo, Plesk and Open Xchange? or does it need more physical ram?

[jmackenz]

Still stuck on this.... The only link that works from the panel is virtuozzo , which looks like it has tons of resources, but still getting nowhere.

Anyone have any advice at all? I'm at a loss here.

Service Temporarily Unavailable
The server is temporarily unable to service your request due to maintenance downtime or capacity problems. Please try again later.


--------------------------------------------------------------------------------

Apache Server at hosting.montellocs.com Port 8443


if anyone thinks they can help i'll gladly privmsg login info.

Currently been paying for this server and the VZ license but getting nothing out of it.

[root@Hosting /]# /usr/local/psa/bin/domain
Plesk is not configured. You should go to the Server Administration Panel and configure it.

But Thats where I get back to issue1

[faris]

I think you should recreate the Container from scratch, this time NOT using the Plesk template. Just create a normal one and give it 512Mb of RAM.

Then manually install Plesk using the latest installer.

But my feeling is that all this has something to do with the Service container more than the container running Plesk itself.

Faris.

[jmackenz]

Just did as you suggested, same results any other ideas?

or would I be better getting an X64 box with more RAM?

[jmackenz]

So I decided to stop procrastinating over the inevitable and had 4gb ram added to my system, and ordered a re-image to 64bit centos.


Will let you know what happens on the rebuild.

[faris]

Just make sure your Service container works correctly once all this is done, and if need be to make sure that the firewall on the container is set to allow the IP of the Service container (although this is now a private IP routed via the Hardware Node itself, so there should not be an issue)

Faris.

[jmackenz]

Now when you say service container, what exactly are you refering to?

In my mind there was the hardware node (the physical server?), and within it were three containers, my open xchange, the attempted plesk, and the PSA Manager

[jmackenz]

for the record, my hardware node resources showed my ram as maxed out.. despite the containers looking as though they had resources to spare.

Maybe I added 1 + 1 and got 3 , but I made the assumption that this was the root cause of my issue...

[faris]

It is common (on VZ) to see the memory maxed out. It is normal and usually not an indication of a problem. If your swap also gets used to max, then there's a problem and the node will slow to a crawl.

VZ4.6 is a little different to 4.0 (which I use).

On 4.0, the service container is a slightly hidden container that does all the background work. Backup, logging in to plesk, PIM, offline management and so on (when you login to the power panel, you are logging in to the service container). It is container number 1 (always). It get allocated a random private IP and only accessible from the hardware node.

On 4.6, you have to install it manually or something. I don't recall the details.

Faris.