Starting from v2.11 Resilio Connect allows restarting in an active job run and in synchronization jobs on an Agent.
Restarting the agent is useful for:
- fixing Database error
- fixing Share's identifying errors
- fixing Missing folder errors
- forcing the agent to re-download data in distribution jobs without restarting the whole job run
It is possible to restart a Synchronization job, a completed or an active job run of Distribution, Consolidation or Script job. The latter will spawn a new job run.
In MC v3.3 the option is not available for Reference Agent in a Synchronization job.
Restarting a synchronization job or an active job run for an agent means that it will:
- delete database (.db, files.db and sf.db) files associated with the job run and create new ones. If there are a lot of files and/or a lot of agents in the job, it will take time for the restarted Agent to scan the files and merge the folder tree with others, days for tens of millions of files.
- delete .sync/ID and part files .!sync. New .sync/ID will be created. The partially downloaded files will be downloaded anew.
- preserve IgnoreList, StreamsList, Archive and other files inside .sync apart from those mentioned above.
This action will produce two events in the job run - agent was removed from the job and added back
Restarting a completed job run for an agent has some peculiarities. If only destination agent(s) is selected, all sources will be automatically added to the new job run. Same if only source agent(as) was selected - all destinations from the job will be added to the new job run.
Full reindexingAs a side effect the agent will have to build indices for all files within job from scratch. If your job contains millions of files that may take significant time to perform reindexing.
The Agent will remain in job run even if it takes long to restart. While the agent is being restarted it will report status 'loading' in the job run.