- When wrong number of bytes is returned in a block read, assume
that initial file-size estimate was different than the current
remote file -- purge the file.
- Make two cinfo checksums: one for the core data and another for downloaded
state and for access statistics. The logic here is that if core part gets
corrupted, I could get nonsensical/wrong values for sizes of the following
arrays and thus I thought it's better to check that first.
- Add checks for corrupt / nonsensical access stats in previous cinfo file
versions.
- Remove support for cinfo file version 1.
- Do not allow enabling of hdfs mode during configuration parsing:
this functionality has not been supported for a while and there is
now a better way of potentially re-implementing it.
You can view, comment on, or merge this pull request online at:
https://github.com/xrootd/xrootd/pull/1349
-- Commit Summary --
* [XCache] Improve cinfo consistency, purge file on read size mismatch.
-- File Changes --
M src/XrdPfc/XrdPfc.cc (10)
M src/XrdPfc/XrdPfc.hh (6)
M src/XrdPfc/XrdPfcCommand.cc (4)
M src/XrdPfc/XrdPfcConfiguration.cc (9)
M src/XrdPfc/XrdPfcFile.cc (21)
M src/XrdPfc/XrdPfcIOFileBlock.cc (4)
M src/XrdPfc/XrdPfcInfo.cc (314)
M src/XrdPfc/XrdPfcInfo.hh (77)
M src/XrdPfc/XrdPfcPrint.cc (5)
M src/XrdPfc/XrdPfcPurge.cc (19)
-- Patch Links --
https://github.com/xrootd/xrootd/pull/1349.patch
https://github.com/xrootd/xrootd/pull/1349.diff
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/xrootd/xrootd/pull/1349
########################################################################
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
