Portal Error - The portal site is currently unavailable. Please contact your portal administrator to resolve the issue (Log files attached)

4657
11
Jump to solution
08-22-2018 04:37 AM
Highlighted
New Contributor

Hello!

I've recently inherited a  portal configuration (version 10.6). The apps seemed to be running fine yesterday; however, now they are down and I am unable to access the portal homepage.

I've attached the .log files under <install dir>\framework\runtime\tomcat\logs folder - as far as i can gather the problem seems to be outlined here:

'Cannot connect to database: jdbc:postgresql://localhost:7654/gwdb Connection refused. Check that the hostname and port are correct and that the postmaster is accepting TCP/IP connections.'

Any help would be appreciated. I'm a long time ESRI user, first time portal admin! 

Reply
0 Kudos
1 Solution

Accepted Solutions
Highlighted
Esri Frequent Contributor

The portal logs on disk should have some errors about whether it's trying to start the database and whether it succeeded or not. Also, you can check the Event Viewer to see why PG won't start. There will likely be lots of noise in there about PG errors, but if PG can't start, one of them will have a clear error and the others can be ignored.

View solution in original post

11 Replies
Highlighted
Esri Frequent Contributor

Have you tried to restart the Portal Windows service? Do you see postgres processes on the Portal machine? It's likely that Portal's internal database isn't running.

Reply
0 Kudos
Highlighted
New Contributor

Hi Jonathan,

Thank you for the response. I did try resarting the service, and the entire server with portal installed. I am still having the same issue however.

What I can't see when I look in the task manager are any postgres processes.

Reply
0 Kudos
Highlighted
Esri Frequent Contributor

The portal logs on disk should have some errors about whether it's trying to start the database and whether it succeeded or not. Also, you can check the Event Viewer to see why PG won't start. There will likely be lots of noise in there about PG errors, but if PG can't start, one of them will have a clear error and the others can be ignored.

View solution in original post

Highlighted
New Contributor

I checked the Event Viewer as you suggested and it turns out PG wouldn't start due to insufficient space on the device. I contacted the IT department to add some more and everything is working now. Thank you very much for the help!

Reply
0 Kudos
Highlighted
Esteemed Contributor

Are there files created with Portal that get saved to a server that would periodically need to be deleted by storage space does not get filled up?  If so, where would these files exist?

Reply
0 Kudos
Highlighted
Esri Frequent Contributor

Unfortunately, yes, there are a few places were files can grow unbounded. We're working on tying up those loose ends:

1) Database logs under the logs directory

2) DBxxxx folders under the default arcgisportal directory that get created during failover or import site

3) walarchive logs in the backup directory that can grow unbounded once you run the DR tool. They are cleared out when you create a full backup

4) Service logs, (the Windows service logs), but this is fixed as of 10.6

Highlighted
New Contributor

Hello jonathan,

i'd assume that we are having similar issue and the portal is not working , when checking the event viewer logs it recorded that PG doesn't have enough space when checking drive C we found that it had 26 MB available after some search e found that walarchive folder consumes around 70 gigs. Accordingly we freed up some space by deleting the walarchive folder files and then restarted the portal server but with no hope. Additionally we did portal for arcgis software repair but withno luck either.please find below the system specs:

- portal for arcgis 10.6.1

- windows server 2012 R2 

 i am also attaching the log files of the internal DB , portal for arcgis logs and tomcat logs for your reference 

Any help would be appreciated!

Reply
0 Kudos
Highlighted
Esri Frequent Contributor

Seems like there are significant problems with the internal database and possible other files:

2019-07-17 15:27:35 PDT: [4340]: LOG:  archive command failed with exit code 1
2019-07-17 15:27:35 PDT: [4340]: DETAIL:  The failed archive command was: copy "pg_xlog\00000001000000250000007E" "C:\arcgisportal\backup\walarchive\00000001000000250000007E"
2019-07-17 15:27:36 PDT: [4340]: FATAL:  archive command was terminated by exception 0xFFFFFFFF

<Msg time="2019-07-14T03:16:06,432" type="WARNING" code="217065" source="Portal" process="1536" thread="1" methodName="" machine="" user="" elapsed="">Error checking status of web server. java.util.concurrent.ExecutionException: org.xml.sax.SAXParseException; lineNumber: 129; columnNumber: 9; XML document structures must start and end within the same entity.</Msg>

Are you taking scheduled backups or snapshots of the environment?

Reply
0 Kudos
Highlighted
New Contributor

Hello Jonathan,

Thanks for your usual support , I managed to solve the issue with the guidance of ESRI support . The server.XML was corrupted due to insufficient space and once we replaced it with a fresh one the portal came to life.

Thanks

Islam Abdel Aal

Reply
0 Kudos