No console output if crawling two types of connectors at the same time

[dagurval]

If you start crawling two collections at the same time (different connectors), the collection that was started first will stop providing console output.

[runarbu]

I can confirm that this is happening to me also, using the newest version from git. Output from the crawler is supposed to be logged to /tmp/coll_log_COLLNAME, but it appears that this breaks when two crawlers are running at the same time.

I will investigate.

[runarbu]

Sorry, I was mistaken about being able to reproduce your issue. I instead observed another issue, where nothing was logged to the console when I forced a new crawl. After some investigation it turned out that the lack of console data was caused by the indexer still working on the previous crawler run, and thus holds a lock preventing me from starting a new crawl of the same collection. I have added a patch so the ES now show "running indexer" as status when the indexer is running ( db1816d ).

Anyway, I am unable to replicate this, having tried to with both the production version and with the newest source code from GitHub.

• Are you using the latest source code from GitHub or the production version?
• Can you please log into the console and do a tail –f /tmp/coll_log_* see if the output from the crawler get logged into the log file. If it does, this may be a gui or browser related issue.