Automatic Write Lock Renewal Failed

[dpatte]

I am importing, processing, and storing images that I am migrating from another site (Drupal) on a different server to a site built with Meteor 1.3.4.1.

Everything seems to work fine until about a minute after all the images are migrated at which point I start getting the above error messages. Since I am importing and processing several thousand images, its not trivial to determine if all the images were actually imported, though I believe they were. Same effect when running locally, or after meteor build and deploy.

Its this err message critical? Should I ignore the messages. It seems as though all (or most of) the images have been processed.

I am using request to retrieve the images, gm to process the images, and FC upsertStream to store the images.

Thanks for your input

[vsivsi]

Hmmm. You shouldn't be seeing this error, because I can't think of any reason why lock-renewals wouldn't routinely just work. But in any case it should only really be possible for this to happen when it is taking a long time to write files to the gridFS/file-collection (for whatever reason).

One thing you can try if this is an ongoing issue is to increase the default lockExpiration time, as described here:

https://github.com/vsivsi/meteor-file-collection#create-a-new-filecollection-object---server-and-client

[dpatte]

Thanks. I may try that. In the mean time, since I'm importing these files in a loop, I decided to try adding a 200ms timeout between files and the issue is gone. Searching elsewhere, it seems I may have run out of file descriptors (?).