Hi we have Portal for ArcGIS 10.4 installed.
On configuring the web adaptor with the portal, it comes up with error message attached.
It seems that Portal is no longer listening on port 7443. Portal has been configured with HTTPS
This was working correctly before, hence we would want to find out what's caused this issue, rather than simply reinstalling portal.
This seems to be similar to issue reported in
Did you follow the troubleshooting steps that were taken in that thread? If you can't reach the portal through the 7443, (ex https://portal.domain.com:7443/arcgis/home/), then you can try to restart the Portal service. When you stop the service, make sure there are no Portal processes running within the Task Manager. Start the Portal service again and see if you can reach the url above. If not, you can try to restart the whole machine, but I doubt that'll do much, and worst case, uninstall and reinstall Portal.
Yes I followed the troubleshooting steps mentioned in other thread. I also stopped the Portal for ArcGIS service and restarted the server, but it made no difference. This was working few weeks ago. Before we go and uninstall and reinstall portal, any idea what could be causing the portal to not respond on port 7443?
We actually have set up our 2 portal servers for High Availability, and also configured a federated/hosting ArcGIS server.
The second portal was joined to the first portal site as per the documentation. The second portal still works correctly on port 7443. When looking at the status of the machines I get following error message for the first portal machine:
Site is not ready yet
The portal postgres log files seem to contain some error/warning messages (see attached)
Hm, which machine is configured to be the primary? The one that is up or the one that is down? Are the logs you provided from the machine that is down? The second portal should be the primary. Do you see 2 javaw.exe processes running and some postrges processes running on the machine that is down? Can you sign in and do all normal stuff on the machine that is up?
The first machine (01) was installed and configured first. This is down currently. The logs are from the same machine (01)
Yes there are 2 javaw.exe and some postgres processes running on this machine.
The primary machine listed is the second machine (02)
I can log on the portal directly via port 7443 and also via webadaptor on this machine.
I can also log on the portal via webadaptor on machine 01.
I'd imagine the reason why you can sign in with the web adaptor on machine 01 is because it sees portal 01 as down and sends it to portal 02. I think you made a mention to it, but just to be clear, is your portal https only, or can you reach portal through http?
Yes our portal is configured for HTTPS only. When I try to access the portal on 01 via http, it works by redirecting to our 3rd party NLB on HTTPS. I'd imagine the traffic then goes from NLB->WA on either 01 or 02 -> portal on 02
1. Is there any portal config file with the port number or SSL settings, which we can look at?
2. Are there any registry settings we could check?
3. Could the issue be cause by environmental factor i.e. issue on server, in our case Windows 2012 R2
4. Does this issue relate to the SSL certificates we have imported into portal?
5. Since portal 02 was initially joined to 01 for HA, would a uninstall and reinstall of portal on 01 cause any problem. How would we go about reinstating the HA of the 2 portals after re-installing portal 01, if this is the only option?
6. Would a reinstall of portal 01 cause us to lose our portal configuration ?
Sorry for the late response. Are you still working on this? To answer your questions:
1) unfortunately there are no config-files or registry settings that can be tweaked/looked at for the SSL settings and ports
2) I wouldn't think this is an environmental issue with the OS, but it could be. Have there been any updates, changes or patches applied?
3) Did you import certificates to be used by 7443 or did you import certificates to be trusted by Portal?
4) There should be no problem unregistering portal 01, reinstalling, and re-registering it. If portal 02 is the primary machine currently and you unregister portal 01, portal 02 simply becomes a standalone portal, retaining all of the settings, content, etc it always had. Once you uninstall and reinstall portal 01, you join it to portal 02 and it will inherit all of the settings, content, etc that portal 02 has, and you're back to an HA configuration.
I'm not sure what could have caused this, though. Had you seen anything relevant in the portal logs on portal 01?