XF 2.0 vB3.8.1 to XF2 import stalled - How to find import error?

[tourmeister]

I started my import from vB 3.8.1 to XF 2 last night at 10:00pm. My database is roughly 2GB with 1.5M posts and around 21K users. When I went to bed at 2:30am, the import was at 50% on step 16 (posts I think) and still running. I checked it this morning at 11:15am before leaving for work. It was stalled at 63%. the little three bar thingy in the upper right corner was no longer moving, nor was the bar that slides across the top from left to right. I left it as it was and went to work, hoping against hope it might still be running...

[edit] I believe it is hung on step 16.

If it is in fact hung, how do I track down the error? Is there a log file somewhere I can view? I've read the import manual several times and cannot find anything describing what to do in the event of an error.

I am running on a VPS. I have access to the WHM and cPanel (although I am VERY new to using both). Using the cPanel File Manager I was able to find a /logs directory with a few files ending in .gz. I tried viewing them with Notepad, but they are unformatted and very hard to make any sense of.

Just looking for some direction on where to go next?

 

[tourmeister]

Okay, I am not 100% sure I have the right logs, but I found these:

/var/log/apache2/error_log and /var/log/apache2/modsec_audit.log

Both files start at Oct 22 10:32, which is when I started the import process (I cut everything before that because the files were HUGE).

I've got no clue what they might be telling me beyond there is something going on with Medsec... whatever that is.

 

[Paul B]

Mod sec is a server security mechanism.

I haven't looked at the logs but it sounds like a rule may be being triggered by the importer.

Ask your host to disable it or whitelist whichever rules are being tripped.

 

[tourmeister]

Would I just need to disable while doing the import or leave it disabled to run XF2 going forward?

 

[Paul B]

It depends whether any other rules are triggered during normal operation.

Typically it's just one or two rules which need sorting out and in most cases it can remain enabled.

 

[tourmeister]

From the Hit List I found in WHM, here are the Rule Id's being triggered:

900095 - Bad UA: Fake Mozilla Attack
980130 - Inbound Anomaly Score Exceeded
949110 - Inbound Anomaly Score Exceeded
920450 - Header is Restricted by Policy
920210 - Multiple/Conflicting Header Data Found
999995 - PHPMyAdmin Script Attack

Would it make a difference if I were doing the import via the CLI instead of the browser? If not, then should I disable each of these rules for the import and then re-enable them afterward?

 

[eva2000]

Maybe ask web host to whitelist your server and your ISP IP addresses you are running the import from https://malware.expert/tutorial/how-to-whitelist-ip-address-with-modsecurity/

 

[tourmeister]

Well, I mentioned that I did not abort the import before leaving for work. At that point, it had been running for 13 hours. When I got home this evening, it was sitting at the screen where it shows time break downs for the various steps and then you click the "finalize import". Interestingly, the time it showed was only 7.6 hours...!? After I hit the "finalize button" it ran for maybe another 30-40 minutes and finished.

As far as I can tell, most everything is present. Some of my forums show as "Private" for some reason and I cannot figure out why. I am sure it is a permissions issue.

I guess now I spend the time it takes to familiarize myself with the Admin CP and its features.