| OHDSI Home | Forums | Wiki | Github |
Followed process as normal using athena web interface. Sent an email with a link that redirected to the following:
http://91.225.130.23/vocab_files/vocab_download_v5_%7B901B08F7-C207-AD9B-C922-1C33569173B8%7D.zip .
The download completed (quickly), but the resulting file failed to open (stuck in a zip <=> cpgz loop).
Tried with both Safari and Chrome. Using Mac OSX Mavericks.
Will fix shortly. Try again.
Fixed
My latest 2 file downloads were both 0 in size (one was to a different, work email address)…
Same here unfortunately. I also tried 2 different email addresses, same result.
My download file was empty also.
Don
Friends:
Sorry about that. It happens when too many people are getting a download and the disk gets full. We are not blaming you, just explaining. The cron job will remove files after a week, but sometimes it will get packed. We need a bigger disk.
Will fix.
Should I resubmit my request, or will you be regenerating the zip file? Should I wait until
you give the “all clear” ?
Thanks,
Don
All clear. Yes, please resubmit. The old submission is gone. We will increase the disk size so we won’t run into this.
I believe I’ve just run into this problem again. Just pulled down a 0-byte zip file from the link I got in an email.
I tried again today, and this time got a 58MB file (which seemed too small, since most other vocab downloads have been in the 200+MB range). When I attempted to unzip the file I got this warning:
End-of-central-directory signature not found. Either this file is not
a zipfile, or it constitutes one disk of a multi-part archive. In the
latter case the central directory and zipfile comment will be found on
the last disk(s) of this archive.
Which is gibberish for “this file has been truncated somewhere in the middle and is broken”. So I’m assuming this is a variation on the “disk is full” problem.
Please fix ASAP.
Friends: Will put a big disk in. Give us a day.
Still having the problem. Anything I can do to help? Where’s the source that drives Athena?
Try now.
Still got a 0-byte file. I’m trying to download all vocabularies possible. Perhaps there is a vocabulary I’m requesting that is causing the build process to fail? I’ll try downloading only the vocabularies that are selected by default and report back.
Nope. Still a 0-byte file, even if I opted for the default set of vocabularies.
On the off chance the problem was related to the v5.0 vs v4.5 vocabularies, I tried downloading the default set for both and both came back as 0-byte files.
The problem is very simple: For some reason, the tool can’t write the zip file. usually, that happens when too many users are downloading simultaneously. But we created enough space. Some stupid permission or so. Should be fixed tomorrow.
I was able to download one file that actually unzipped back on Friday night. It was version 5 with the standard set of vocabularies.
I tried downloading all available vocabs for 4.5 and 5.0 and both downloads gave me large, 1.08GB zip files that failed to unzip with the same error I mention above.
I tried downloading vocab 5.0 with the standard set since then and it’s failing again (I’m getting 0-byte files).
Today I was able to download the default set of vocabularies for version 5. However, if I select all the available vocabularies for version 5, I get a 1.0GB file that fails to unzip.
Also, in the past I would occasionally submit several requests to Athena within a few minutes of each other. It seems like it takes Athena A LOT longer to complete the build and send me an email with my link when I do this. Also, my downloads seemed to fail (either 0-byte files or bad zip files) when I would do this. Just FYI. I imagine that I was putting a lot of load on Athena by making multiple submissions, but queuing theory would suggest this is a real-world scenario that Athena should be able to handle.
I’ll work on downloading a 4.5 default and full set of vocabularies tonight and report back later.
Thanks!