Unable to launch rsync daemon on backup host

The backup on the vertica host fails after the upgrade to 7.2.2-2 with the below error:  

 

Warning: the config parameter 'verticaConfig' has been deprecated.
Error: Unable to launch rsync daemon on backup host:****.

 

 

Comments

  • Hi
    Try menualy start the rsync server on the target hosts using this metod

    etc/initd/rsync start
  • /etc/initd/rsync start
  • We had the same errors after upgrading to Vertica 7.2.

     

    The first problem just refers to an option in the config file that is no longer supported, so we just deleted that line. We were using the previous default anyway, so I don't think it will cause any problems for us.

     

    Regarding the second problem, we tried starting the rsync daemon manually, as suggested here, but that did nothing. Using ps, we were able to see that an rsync daemon was in fact running, but it was Vertica's own installation of rsync, the one found in /opt/vertica/sbin. It seems that the upgrade caused a miscommunication between Vertica and its rsync daemon, so that Vertica couldn't see that it was running, and also couldn't start it.

     

    Our solution was to kill the rsync daemon manually with pkill, and then re-initialise Vertica's backup configuration with

     /opt/vertica/bin/vbr.py --config-file [name_of_config_fil].conf --task init

     

    After that our backups were able to run.

  • Thanks gayle for the reply and the solution, we had to follow the same process to get the backups working. Did your issue come back the next time the backups ran?

  • No, the backups ran fine after that. The only other issue that we came accross was that the backups were stored in a different way to before the upgrade, so instead of storing one copy of the database (and overwriting the previous copy), plus snapshots, we were actually trying to store 2 copies, one from before the upgrade and one from after, which caused us to run out of space on our backup disk. So we had to delete the old copy, and now everything seems to be fine.

     

     

  • Yes, we had to overwrite the backups as well.

  • I can confirm the same issues (and solutions) as above. 

     

    Constantly impressed at just how awful this software is. 

     

     

  • Another thing to try: check that the uid and gids for both dbadmin and the verticadba groups are consistent between the database nodes and the backup hosts. An inconsistency in the verticadba group also gave us this error.

Leave a Comment

BoldItalicStrikethroughOrdered listUnordered list
Emoji
Image
Align leftAlign centerAlign rightToggle HTML viewToggle full pageToggle lights
Drop image/file