 |
 |
> When we were doing ATLAS that was an enduring problem for certain DPM sites. Right, that was my first thought. However, my recollection was this was very reproducible - they basically advertised the existence of every file. This happens with the probability of something like `10^-5` (which also makes it extremely difficult to figure out where things are coming from). > Can we narrow this down? For instance, can we get a repeatable case? Unfortunately, I've been trying for about 2 days to come up with something repeatable - or at least seeing this in the lab. No luck :( > Finally, do we have an idea of what client version is involved here A number of versions of the client are used, depending on the software release. If you want to zero in on one, 4.8.5 is perhaps the most common. > Is it possible that there is a problem in the software framework that drives all of this? Certainly - I haven't been able to rule this out. However, the problem with this theory is that file names X & Y are sometimes completely different - different datasets, different software versions, different production "campaigns" and for the most part not mentioned in the same job. We have only found one case where X & Y were both input files in the same job. We are still waiting for redirector logs on this. Depending on the redirector log contents, we should really be able to narrow down the suspects. -- You are receiving this because you commented. Reply to this email directly or view it on GitHub: https://github.com/xrootd/xrootd/issues/1097#issuecomment-561776061 ######################################################################## Use REPLY-ALL to reply to list To unsubscribe from the XROOTD-DEV list, click the following link: https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=XROOTD-DEV&A=1