[
https://issues.apache.org/jira/browse/MINIFICPP-303?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16251683#comment-16251683
]
marco polo edited comment on MINIFICPP-303 at 11/14/17 4:42 PM:
----------------------------------------------------------------
[~calebj] Your comment in 262 was "I can't reproduce the issue either on a
fresh Trusty VM," but yet you were having issues on Cloud9. Did this solve the
Cloud9 issue for you ? I saw your comment, "It might solve." I don't want to
close 262 until I can verify from [~achristianson] that 262 was a problem he
could/can replicate and is still a problem.
Sorry I said, "before merging," but I don't plan to hold merging this to get an
understanding.
was (Author: phrocker):
[~calebj] before merging 303 I want a handle on understanding what the problem
was in RocksDB. Your comment in 262 was "I can't reproduce the issue either on
a fresh Trusty VM," but yet you were having issues on Cloud9. Did this solve
the Cloud9 issue for you ? I saw your comment, "It might solve." I don't want
to close 262 until I can verify from [~achristianson] that 262 was a problem he
could/can replicate and is still a problem.
> Upgrade civetweb and rocksdb
> ----------------------------
>
> Key: MINIFICPP-303
> URL: https://issues.apache.org/jira/browse/MINIFICPP-303
> Project: NiFi MiNiFi C++
> Issue Type: Improvement
> Reporter: Caleb Johnson
>
> These upgrades to the latest stable releases of civetweb
> ([v1.10|https://github.com/civetweb/civetweb/releases/tag/v1.10]) and RocksDB
> ([v5.8|https://github.com/facebook/rocksdb/releases/tag/v5.8]) bring numerous
> improvements. They might solve any issues similar to those in MINIFICPP-262.
> For civet, the most notable is OpenSSL v1.1 support. It was partially
> backported to the version in the minifi tree (1.9.1), but still had issues
> matching some OpenSSL v1.1 interface changes. Civetweb release v1.10 has
> complete and official support.
> RocksDB v5.8 comes with the following bugfixes:
> * Fix wrong latencies in rocksdb.db.get.micros, rocksdb.db.write.micros, and
> rocksdb.sst.read.micros.
> * Fix incorrect dropping of deletions during intra-L0 compaction.
> * Fix transient reappearance of keys covered by range deletions when memtable
> prefix bloom filter is enabled.
> * Fix potentially wrong file smallest key when range deletions separated by
> snapshot are written together.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
