EMS 7.4.4 Web Server 500 Error after migration

[SteveJW]

In my test environment, I migrated from FortiEMS client 7.2.2 to 7.4.4.
We deployed EMS on a Hyper-V VM.
I didn't encounter any issues during the migration.
I want to start the console now, but I'm getting the error below.

I think it has something to do with the certificate, among other things.
How can I remove it and/or configure/import a new certificate?

[SteveJW]

Hi AEK,
See printscreen

[AEK]

I see it is configured on the default port 5432, so I wonder why the app is trying to connect to 6432 instead?

I don't know how this could happen, and I don't find a way how to tell the app to use 5432 instead of 6432, but as a workaround I'd suggest to try change the DB port to 6432 so the app can find it with its current setting.

config set db --db.port 6432

The reboot the VM (probably required).

In case it doesn't work then rollback the change like this:

config unset db --db.port

Hope it helps.

[SteveJW]

Unfortunately, changing the DB port to 6432 didn't solve the problem.

I'll run the migration again.

Fortunately, I created a snapshot, so I can start the migration in a few minutes.

I only ran certain commands after the migration, such as configuring a static IP address and the timezone.

I don't think that's the cause, but I'll run the following before migrating:

system set network ip --adapter="eth0" --ip="10.10.10.22/24" --dns="10.10.10.1" --gateway="10.10.10.1"

execute timezone set Europe/Amsterdam

Are there any other commands I can run before starting the migration?

For example, I see that updates can be installed.

How can I install the updates?

[AEK]

These updates are Linux related. I didn't use the virtual appliance version yet, so I don't know how to manage Lunix updates in this new version. Before it was possible because Linux shell was directly accessible.

Hope some more experienced member can help.

[SteveJW]

I reran the migration and can now access the console.

The only things I've changed this time are entering static IP addresses and configuring the timezone for the migration.

Later, I noticed in the DHCP leases that the Linux EMS server, during the previous migration, initially received a different IP address.

After a reboot, it received a different IP address again.

This could maybe have caused the web access issues.

Anyway, in the console output, I saw percentages and verifications that didn't show 100%.

I also had this during the first migration test
I don't know if this still requires further attention.

See link for migration log file

https://filebin.net/4yvvgrobkgkublaq

See link for "all log files"
https://filebin.net/y2qgdmgd0zsikuzv

If applicable, I'd appreciate recommendations.

If not, I'll close this post after a few days.

I'll now review the configuration and start testing the client migration.

So, you won't be rid of me anytime soon :)

[AEK]

Happy to hear that the migration worked this time.

I can't comment the failure messages on the output, nevertheless since we can see "EMS Migration Successfully completed" and since there is no fatal error then for me I'd consider it done successfully.

If you want to be sure for your production, you can still open a ticket with these messages to get feedback from Fortinet TAC.

[SteveJW]

For production, I want to be sure I don't run into any vague issues.
I've created a ticket.
I'll let you know when they respond.

[gnk48485]

I ran into the same issue after upgrading FortiClientEMS from 7.4.3 to 7.4.4. The web interface now shows a 500 Internal Server Error.

Looking at the logs, I frequently see this error:

ERROR manager/socket_pool.go:212 Failed to create new socket connection: dial tcp 127.0.0.1:65432: connect: connection refused

Has anyone else encountered this after the 7.4.4 upgrade? Any ideas on how to fix it?

[SteveJW]

It's strange that you're getting the same error during an upgrade.
After setting a static IP address and time zone, the problem was resolved after a second migration test.
How do you connect to the EMS server?
Try the FQDN or IP address.
Is the time zone correct?

I have absolutely no Linux knowledge.
But with the migration to this version, i do need to learn some Linux/emscli commands.

[gnk48485]

timezone is correctly set, the static IP is configured and correct, I normally access EMS via FQDN with a valid certificate, and I tested direct access via IP address as well — the error remains the same.