Joe McDonnell has posted comments on this change. Change subject: IMPALA-4623: Enable file handle cache ......................................................................
Patch Set 11: (4 comments) http://gerrit.cloudera.org:8080/#/c/6478/11/be/src/runtime/disk-io-mgr-scan-range.cc File be/src/runtime/disk-io-mgr-scan-range.cc: Line 445: // bad file handle; record the error, but allow for a retry to fix it. > wouldn't the same apply to the pread path? Added retry for hdfsSeek in Open() and hadoopReadZero in ReadFromCache. Since there are multiple locations for retry, I created a separate function that reopens the file handle. Changed the hdfsPread/hdfsRead paths to fill in status inside the loop. Moved this comment up top and made it apply to all the operations. Line 467: if (current_bytes_read != -1 || num_retries >= 1 || > num_retries == 1 and dcheck that it's not >1 somewhere? Done Line 468: (hdfs_file_from_cache == nullptr)) { > remove ()? Done Line 471: // If the file handle is from the cache, it might have become invalid. Destroy > rephrase, we know it's from the cache Done -- To view, visit http://gerrit.cloudera.org:8080/6478 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-MessageType: comment Gerrit-Change-Id: Ibe5ff60971dd653c3b6a0e13928cfa9fc59d078d Gerrit-PatchSet: 11 Gerrit-Project: Impala-ASF Gerrit-Branch: master Gerrit-Owner: Joe McDonnell <joemcdonn...@cloudera.com> Gerrit-Reviewer: Dan Hecht <dhe...@cloudera.com> Gerrit-Reviewer: Joe McDonnell <joemcdonn...@cloudera.com> Gerrit-Reviewer: Marcel Kornacker <mar...@cloudera.com> Gerrit-Reviewer: Tim Armstrong <tarmstr...@cloudera.com> Gerrit-HasComments: Yes
