Pause indexing problem

[firstrebel]

If I pause indexing (build 1011) for a few minutes, when I resume I get error messages as the files that were being downloaded at the time I paused could not be indexed.

This has crept in recently, it never happened before but I can't remember which version I last used when it was OK.

This is linked to the time that indexing is paused, but I have not tried to find out where the line is.

Bob

[Ray]

This is probably the same as it's always been, but it would depend on whether or not you happened to pause during a particularly big file, or for a particularly long time, or if your server has a particular timeout period (come to think of it, the latter is probably the most likely, in that the timeout period may have been reduced on your server so it's more likely to hit that timeout now).

Basically, because we allow the user to pause the actual downloading immediately (rather than let it continue/finish downloading before pausing), it is possible that you can pause the download for too long a period, and the server will refuse the continuation of downloading of that file. Zoom does not send a second HTTP request when you resume. Although this is probably something we should consider for V6.1 or similar. Since many servers do not support HTTP resume, we may even need to re-start the entire download for that file, so there's a fair bit of change to handle that scenario.

[firstrebel]

I specifically remember with some previous version the download(s) taking place finished before the process paused.

Bob

[Ray]

When you "Stop" the indexing, it will finish downloading before proceeding to stop completely. But "Pause" always paused the downloading immediately (I even just checked in V5.0 and V5.1 to be sure).