[SOLVED] Server Status connection refused in admin console

[tkosig]

I know there has been a similar thread before that has been marked solved (here). The author of the thread solved his problem by complete re-installation of the ZCS which at this time is NOT an option to me.

We're in the process of evaluating ZCS Network Edition and so far we like it. The problem I need help with is that accidentially I installed the zimbra-logger package on both mailstore and MTA server.

I quote from the other thread (as I am having exactly the same problem!):

When I click on the server status item in the left panel of the administration web page, I received the following error:

Message: error while proxying request to target server Connection refused Error code: service.PROXY_ERROR Method: GetServiceStatusRequest Details:soap:Receiver

In the mailbox.log file I see:

2009-08-18 09:09:39,192 INFO [btpool0-34] [name=timo.kosig@mycompany.com;mid=4;ip=192.168.100 .29;ua=ZimbraWebClient - FF3.0 (Win);] SoapEngine - handler exception
com.zimbra.common.service.ServiceException: error while proxying request to target server Connection timed out
ExceptionId:btpool0-34:1250579379192:d84fee73f357f6d5
Code:service.PROXY_ERROR Argurl, STR, "https://mta.mycompany.com:7071/service/admin/soap/GetServiceStatusRequest")



Our sever setup is quite simple. We have a dedicated mailstore server in the trusted network, and the MTA is located in our DMZ.

I disabled the logger on the MTA after realising my mistake. Still it seems that a request is sent to the MTA on port 7071 which times out eventually causing an exception (I left the stack trace out but can supply one anytime needed).

Please help me to fix this problem. As I said before, re-installation is not feasible.

Thanks in advance,
Timo Kosig

[tkosig]

I solved the problem by setting zimbraLogHostname to the right hostname. After installing the logger on the mta zimbraLogHostname was set to mta.company.com.

zmprov modifyConfig zimbraLogHostname mailstore.company.com

Did solve the problem - without re-installing everything.

[ECB]

That solved my problem too. Thanks for the tip!

[sufehmi]

Thanks tkosig ! Your posts helped me solved my problem.

Basically I have accidentally setup a Zimbra server with localhost.localdomain set as its hostname. Very bad.

Some other post suggested to delete it from the Global Server config. However, it failed with error message :

> zmprov ds localhost.localdomain
ERROR: service.TOO_MANY_HOPS (mountpoint or proxy loop detected)

So I checked to see what services that it had registered :

### set variables
su - zimbra
source ~/bin/zmshutil
zmsetvars
### start checking current setting
ldapsearch -LLL -x -D $zimbra_ldap_userdn -w $zimbra_ldap_password -H $ldap_url -b "" | grep -i localdomain
zmprov -l gaa -s localhost.localdomain
zmlocalconfig | grep -i localdomain
zmprov gacf | grep -i localdomain

Looks like zimbraLogHostname is the most important one.

So I found your post, and how to change it :

> zmprov modifyConfig zimbraLogHostname my.zimbra.server

Voila ! No more "PROXY_ERROR Method: GetServiceStatusRequest" error message whenever I tried to access the Global Config

Thanks again !