We have a small Tier 3 xrootd system (a redirector and three servers with comprising about 150 TB capacity, running Copr.  2010 Stanford University, xrd version v3.1.1) and we have a user reporting read errors such as the following:

--------------
I'm getting odd errors like this in my log files:

130228 18:24:14 4767 Xrd: XrdClientMessage::ReadRaw: Failed to read header (8 bytes).
130228 18:27:38 4756 Xrd: CheckErrorStatus: Server [uct3-xrd.mwt2.org:1094] declared: session not found(error code: 3011)
130228 18:32:44 4765 Xrd: XrdClientMessage::ReadRaw: Failed to read header (8 bytes).
130228 18:37:02 4756 Xrd: CheckErrorStatus: Server [uct3-xrd.mwt2.org:1094] declared: session not found(error code: 3011)
130228 18:42:14 6518 Xrd: XrdClientMessage::ReadRaw: Failed to read header (8 bytes).

I'm sorry to say it's not necessarily reproducible. Sometimes a file will spit these out when run over, sometimes a different one will. When this does happen it takes about 50x as long to run over the files, and the output ROOT file with histograms and a small TTree is 1GB instead of the usual 5MB.
---------------

I've been told previously that "ReadRaw" errors are harmless, but hit appears not to be the case in this instance. Any idea what this might be gong on?

Thanks and cheers,

---
Rob Gardner • Skype rwg773 • 312-804-0859 • University of Chicago


Use REPLY-ALL to reply to list

To unsubscribe from the XROOTD-L list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=XROOTD-L&A=1