Hi there,
I am having a problem after I submit my job, I get an error message that says MD5_FILE_CHECK_FAILED.
Looking at the files, it was the first file in the first dataset that caused the problem, the error message is "Invalid checksum". I tried removing this file, but it I got the same problem with the next file too.
My job number is 1002782.
I have converted .raw files to .mzXML files and have been able to upload and process similar files previously.
Any suggestions welcome!
Katherine.
We have been updating XCMS Online recently, specifically the data pre-processing modules to use sha1 instead of md5 checksums on certain files. Unfortunately there was a bug in the way it was implemented that forced the system to check your files (stored MD5 checksum in database) with the locally-calculated sha1 checksum. Since these values did not match, the system errored out the job.
I made a correction to the aforementioned check and reran the job 1002782 (with error) but now there is another problem - I only see 35 files in the source dataset #47917. You would need to reupload the missing files or delete them from the dataset as they are not on disk. On a bright note, I added an additional check to the stored datasets page to display a file as inactive if it is not physically on disk. You can then re-upload the missing/corrupted files.
Duane
Hi Duane,
Thanks for your reply. I re-uploaded the problem files and re-submitted the job. Almost all of the files have been fine, which is great. However, I've an error message for 3 of them:
Error in xcmsRaw(file, profstep = 0) : Time for scan 7998 (3600) greater than scan 7999 (0.402)
Is there anything I can do about this? It's only 3 samples out of over 200, but 2 of them in particular are quite important ones, I'd rather not just delete them.
Best,
Katherine.