Marklogic scheduled backups failing

1
votes

We are currently using scheduled backups to have daily database backups. We are observing many problems in taking backups from past few days. The backup process executes for several hours but didn't get completed. We also had a look at the logs, but there is nothing mentioned for scheduled backup. We have following questions -

  1. By any means, can we make it more consistent and stop such things?
  2. Is there a way to get some kind of notification/email after the certain time if scheduled backup isn't successfully completed until that time.
  3. Can we also have a retry mechanism on the failure of scheduled backup?
1
marklogicmarklogic-8
What about the database status page? Does it show the db as still backing up? What 'file log level' are you running at? Some messages will show up at the Info level, others will show up at the Warning or 'error' level. Do you see [Canceling backup] in the errorlog? The messages are logged to the host that the forest is attached to, so if the db backup fails it will cancel all the other associated forest backups - and write the canceling message to the attached host, and the error that stopped the backup will log to its attached host. Do you have sufficient space on backup drive? - Michael Gardner

1 Answers

3
votes

For your points:

1) If you don't describe he failures, there is little people can do to give guidance.

2) The forest backup information is logged. You can use standard log monitoring/alerting tools to manage alerts. Keep in mind that the log messages are per forest, so you may need to look closely at the lg messages for when a 'complete' or 'failed' backup are done.

3) From the result of number (2), you may decide to kick-off a new backup via the management API. See here: http://docs.marklogic.com/REST/POST/manage/v2/databases/[id-or-name]#BackupDB