what happens if a "tahoe backup" process is in progress when another one is started? #2062

New Issue

zooko · 2013-08-20T14:25:44Z

zooko commented

2013-08-20 14:25:44 +00:00

How does the "tahoe backup" process handle the situation that another "tahoe backup" process is running at the same time, accessing the same backupdb?

To answer this question, start here: [docs/backupdb.rst]source:trunk/docs/backupdb.rst?rev=83f97dcf7dc968ee8442bf5caf011a70a8971b90 and [src/allmydata/scripts/backupdb.py]source:trunk/src/allmydata/scripts/backupdb.py?annotate=blame&rev=a1a1b5bf8a6e1a09c70d4d2b618ad0c06b0f631f.

This is relevant to #2053, because a simple, easy improvement for #2053 is just to make sure that backup processes don't corrupt each other's state if they run concurrently, and that they make good quick progress, and take advantage of one another's (complete, consistent) results for faster progress.

How does the "tahoe backup" process handle the situation that another "tahoe backup" process is running at the same time, accessing the same backupdb? To answer this question, start here: [docs/backupdb.rst]source:trunk/docs/backupdb.rst?rev=83f97dcf7dc968ee8442bf5caf011a70a8971b90 and [src/allmydata/scripts/backupdb.py]source:trunk/src/allmydata/scripts/backupdb.py?annotate=blame&rev=a1a1b5bf8a6e1a09c70d4d2b618ad0c06b0f631f. This is relevant to #2053, because a simple, easy improvement for #2053 is just to make sure that backup processes don't corrupt each other's state if they run concurrently, and that they make good quick progress, and take advantage of one another's (complete, consistent) results for faster progress.