Koozali.org: home of the SME Server
Obsolete Releases => SME Server 7.x => Topic started by: AaronG on November 06, 2006, 04:31:05 AM
-
Hi All,
Before I log this as an error in the bug tracker I would like to do some troubleshooting steps.
In the last couple of days I have been unable to access my webmail and/or server-manager.
I used to be able to access the webmail via any IP address using some Port forwarding on my router.
How can I stop/start the HTTP service?
What troubleshooting shoudl I do?
regards
Aaron
-
I should have said I have tried the following things:
1. Run signal-event post-upgrade & signal-event reboot
2. I CAN access the Server-Manager using an SSH session and accessing it via the Server Console. But is this HTTP?
Thanks in Advance!
-
AaronG
>..unable to access my webmail and/or server-manager.
> I used to be able to access the webmail via any IP address
> using some Port forwarding on my router.
Is port 443 still forwarded ie https: ?
-
Is port 443 still forwarded ie https: ?
Yes I believe this is working ok.
I also can not access HTTP from the LAN.
[/quote]
-
AaronG
> I CAN access the Server-Manager using an SSH session and accessing
> it via the Server Console. But is this HTTP?
Yes it is.
Check services with these
/etc/init.d/squid status
/etc/init.d/httpd-e-smith status
/etc/init.d/httpd-admin status
-
Check services with these
/etc/init.d/squid status
/etc/init.d/httpd-e-smith status
/etc/init.d/httpd-admin status
I don't know if it makes any difference but I am running SME v7 (Latest Build)
Squid Status:
/etc/init.d/squid status squid (pid 3337) is running...
e-Smith Status
/service/httpd-e-smith: down 0 seconds, want up
Admin Status
/service/httpd-admin: run (pid 3246) 15639 seconds, normally down
I take it e-smith Status is my problem?? But how can I further troubleshoot this?
I tried to start the service by using: /etc/init.d/httpd-e-smith status
But I still get the same status. It says OK when i try to start but when i view the status it still says "want up" etc
Thanks for your help so far. It is greatly appreciated!
-
AaronG
> /service/httpd-e-smith: down 0 seconds, want up
I believe that is a supervised service, so it should have restarted automatically in the event that it somehow stopped.
Did you do anything to deliberately stop it eg install sme7admin panel & stop web services ?
Is the service status enabled or disabled at start up ?
config show httpd-e-smith
httpd-e-smith=service
TCPPort=80
access=public
status=enabled
> I take it e-smith Status is my problem??
Seems to be.
/etc/init.d/httpd-e-smith restart
or more correctly
/etc/rc.d/rc7.d/S86httpd-e-smith restart
Restarting httpd-e-smith [ OK ]
-
config show httpd-e-smith
httpd-e-smith=service
TCPPort=80
access=public
status=enabled
is what I get. So it appears enabled.
I DO have sme7admin panel installed but I dont believe I ever stopped the service or did anything to disable it.
I tried to restart the service as you suggested but when I check the status it tells me the same.
Cheers
Aaron
-
AaronG
> I tried to restart the service as you suggested
>> Did you get any message, if so what ?
Try to restart the service and then also look in
/var/log/messages
log file for errors
-
> I tried to restart the service as you suggested
Did you get any message, if so what ?
Restarting httpd-e-smith [ OK ]
I can't even find an entry in the sme7admin panel for httpd-e-smith.
System services (Click on a service name for details) :
Web services State Nb process Action
httpd-admin Started 1 stop
squid Started 1 stop
I have admin but no http
I ran a HTTP test using sme7admin... it returned the following error:
Execution of 'http@' (should return html source code of server's
homepage):
tcpclient: unable to connect to 127.0.0.1 port 80: connection refused
Ideas?
-
AaronG
> Restarting httpd-e-smith [ OK ]
That's good, that's what you want to see.
> I can't even find an entry in the sme7admin panel for httpd-e-smith.
httpd-e-smith is web server
httpd-admin is server manager
> tcpclient: unable to connect to 127.0.0.1 port 80: connection refused
Is port 80 forwarded from your router.
Perhaps you should restart your router and/or modem and also reboot your sme server while you are at it.
Check all the settings in your router carefully.
-
> I can't even find an entry in the sme7admin panel for httpd-e-smith.
httpd-e-smith is web server
httpd-admin is server manager
Problem is there doesn't seem to a "Service" listed in sme7admin panel that says httpd-e-smith.
Is port 80 forwarded from your router.
Perhaps you should restart your router and/or modem and also reboot your sme server while you are at it.
Check all the settings in your router carefully.
This is definately not a problem. this has worked for over 2 years and the problem occurs on the LAN as well when there is no PAT (Port Address Translation) interferering with it.
Do I need to log an error in the bug tracker?
-
AaronG
>...this has worked for over 2 years
Is this an sme7 upgraded from sme6. If so did http work after the upgrade ?
> In the last couple of days I have been unable to access my webmail and/or server-manager.
What did you do to your system just before the problem started.
-
It was an upgrade at first... but this was done a very long time ago. but I have done a full clean install since.
I *think* it happened in the past few days. I only really noticed today. But I remember that the last thing that I did before I think it probably stuffed up was a YUM UPDATE which updated user-panel and user-vacation... both contribs by DUNGOG.
Naturally after I installed these updates i ran the Signal-event post-upgrade and signal-event reboot command.
-
AaronG
> YUM UPDATE which updated user-panel and user-vacation
There was a bug in a recent version of user-panel, don't know if that is related to your problem, but possibly it is.
http://forums.contribs.org/index.php?topic=34354.msg147912#msg147912
Try upgrading to the latest version
smeserver-userpanel-0.9-8.noarch.rpm
-
already on version 0.9.9 of user panel
Damn! :( this is soo annoying
-
AaronG
> version 0.9.9 of user panel
I can't keep up with the releases.
If you suspect it's those rpms or any others for that matter, uninstall them
rpm -e rpmname (no number)
rpm -e smeserver-userpanel
etc
Are there any error messages in the log files when you try restarting httpd-e-smith ? That will be your best clue.
-
I will try uninstalling..
I have logged this as a bug in the tracker...
http://bugs.contribs.org/show_bug.cgi?id=2035
I will uninstall and let you know how I go... there are sooo many damn logs lol. which log do I look in? log name=Messages ???
-
AaronG
> there are sooo many damn logs lol. which log do I look in? log name=Messages
Look back a few posts, the answer was provided already.
-
There arn't really any error messages in the /var/log/messages log file
I ran the following commands.
rpm -e smeserver-userpanels
rpm -e smeserver-userpanel
rpm -e smeserver-vacation
rpm -e smeserver-sme7admin
I checked after each one was uninstalled. Still no good.
I then ran, the usual signal-event --- commands. and rebooted
Still no good. I can't access HTTP or HTTPS.
So using SSH and the console>Access Server console> i opened the HTTPD log... I am getting MANY MANY errors. No space on my harddisk??
LOG: HTTPD.log
[Mon Nov 06 22:18:49 2006] [crit] (28)No space left on device: mod_rewrite: could not create rewrite_log_lockConfiguration Failed
[Mon Nov 06 22:18:50 2006] [crit] (28)No space left on device: mod_rewrite: could not create rewrite_log_lockConfiguration Failed
[Mon Nov 06 22:18:51 2006] [crit] (28)No space left on device: mod_rewrite: could not create rewrite_log_lockConfiguration Failed
[Mon Nov 06 22:18:53 2006] [crit] (28)No space left on device: mod_rewrite: could not create rewrite_log_lockConfiguration Failed
HTTPD/Admin_error_log
[Mon Nov 06 22:18:48 2006] [crit] (28)No space left on device: mod_rewrite: cou
ld not create rewrite_log_lockConfiguration Failed[Mon Nov 06 22:18:49 2006] [c
rit] (28)No space left on device: mod_rewrite: could not create rewrite_log_loc
kConfiguration Failed[Mon Nov 06 22:18:50 2006] [crit] (28)No space left on dev
ice: mod_rewrite: could not create rewrite_log_lockConfiguration Failed[Mon Nov
06 22:18:51 2006] [crit] (28)No space left on device: mod_rewrite: could not c
reate rewrite_log_lockConfiguration Failed[Mon Nov 06 22:18:52 2006] [crit] (28
)No space left on device: mod_rewrite: could not create rewrite_log_lockConfigu
ration Failed[Mon Nov 06 22:18:53 2006] [crit] (28)No space left on device: mod
_rewrite: could not create rewrite_log_lockConfiguratio......
-
AaronG
I didn't suggest you do any of that stuff you did.
I did suggest that you restart httpd-e-smith and look at the messages log file, which you still have not done.
Please read what I write carefully instead of glossing over it.
Run this and note the time you do it
/etc/rc.d/rc7.d/S86httpd-e-smith restart
then review the messages log file and look for entries around that time to see if there are any errors recorded
please also check again if that service is running now, what does it say on the screen when you do
/etc/rc.d/rc7.d/S86httpd-e-smith status
> No space on my harddisk??
What does this give
df -h
-
Filesystem Size Used Avail Use% Mounted on
/dev/mapper/vg_primary-lv_root
146G 2.4G 136G 2% /
/dev/md1 99M 32M 63M 34% /boot
none 248M 0 248M 0% /dev/shm
obviously not a problem.
The bugtracker is pretty active at the moment. http://bugs.contribs.org/show_bug.cgi?id=2035
I had done a few steps in there such as posting logs etc. if u are interested you might wanna check it out, so i don't waste your time
-
Here's a post on the No space left on device: mod_rewrite issue: [FIXED] httpd wont start (http://forums.contribs.org/index.php?topic=33171.0)
-
I am confused from that article. what command do i have to run?
There is talk in the one of the last posts about a script. how do i create and execute a script file with the posted contents???
Cheers
Aaron.
-
Hi,
Can someone please help me create and execute the script which is mentioned in this forum article.
http://forums.contribs.org/index.php?topic=33171.0
It sounds like this script will address my problems.
-
AaronG
> Can someone please help me create and execute the script ....
Log in as root and type the commands indicated in the script
-
I typed in the script etc. and executed it all.
All seemed ok. but it didn't solve my problem... the next post explains some extra steps...
Same problem.
Deleting the .crt and .key files, signal-event domain-modify, and running steveo's script (thanks!) seem to have solved the problem.
Does anyone know why this happens?
Can anyone tell me how i delete the .crt and .key files?? I know how to execute the signal-even domain-modify command.... then if I type in the script, it might work.
But what files do i need to delete and how do i do this? Obviously log in with root.. but what directory do I need to browse to and what files do i delete?
THANKS IN ADVANCE!
-
AaronG
> Deleting the .crt and .key files, signal-event domain-modify, and
> running steveo's script (thanks!) seem to have solved the problem.
I think that was summarising the steps carried out rather than adding extra steps.
> Can anyone tell me how i delete the .crt and .key files??
The steps used the scripts and in there it said how to remove those files.
rm /home/e-smith/ssl.crt/darkstar.edenmonarolabour.org.crt
rm /home/e-smith/ssl.key/darkstar.edenmonarolabour.org.key
You will need to cd / to that location and check the name of your files, and use those names instead.
Are you really reading what has been said to do in that post ?
Please read it all carefully again
-
your absolutely correct RayMitchell.
I was browsing over them. It is past midnight here in Oz. I am sorry for wasting your time but i am EXTREMELY appreciative of the help you have given me.
It did fix my problem. Thankyou very much! i have learnt some valuable troubleshooting steps. I live in a "windows world" so every little bit learnt is great.
THANKYOU!
-
AaronG
> I was browsing over them. It is past midnight here in Oz.
I'm in Oz too, so it's equally as late for me and I was still able to read it all (for you) !
> It did fix my problem.
Please add the link to this thread and the results of your fix to the bug report previously lodged.
It will help others and resolve that bug report (hopefully).