[
https://issues.apache.org/jira/browse/SOLR-14498?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17118589#comment-17118589
]
Erick Erickson commented on SOLR-14498:
---------------------------------------
OK, this is going to be harder than I thought. If I run the gradle
udateLicenses, then gradlew check passes but ant precommit doesn't and vice
versa. I'm not quite sure how to resolve, haven't seen this before. Note that
they both produce exactly the same checksum AFAICT.
> BlockCache gets stuck not accepting new stores
> ----------------------------------------------
>
> Key: SOLR-14498
> URL: https://issues.apache.org/jira/browse/SOLR-14498
> Project: Solr
> Issue Type: Bug
> Components: query
> Affects Versions: 6.5, 6.6.5, master (9.0), 7.7.3, 8.5.1
> Reporter: Jakub Zytka
> Assignee: Andrzej Bialecki
> Priority: Major
> Fix For: 8.6
>
>
> {{BlockCache}} uses two components: "storage", i.e. {{banks}} and "eviction
> mechanism", i.e {{cache}}, implemented by caffeine cache.
> The relation between them is that "storage" enforces a strict limit for the
> number of entries (
> {{numberOfBlocksPerBank * numberOfBanks}}) whereas the "eviction mechanism"
> takes care of freeing entries from the storage thanks to {{maximumSize}} set
> for the caffeine cache to {{numberOfBlocksPerBank * numberOfBanks - 1}}.
> The storage relies on caffeine cache to eventually free at least 1 entry from
> the storage. If that doesn't happen the {{BlockCache}} starts to fail all new
> stores.
> As it turns out, the caffeine cache may not reduce it's size to the desired
> {{maximumSize}} for as long as no {{put}} or {{getIfPresent}} which *finds an
> entry* is executed.
> With a sufficiently unlucky read pattern, the block cache may be rendered
> useless (0 hit ratio):
> cache poisoned by non-reusable entries; new, reusable entries are not stored
> and thus not reused.
> Further info may be found in
> [https://github.com/ben-manes/caffeine/issues/420]
>
> Change in caffeine that triggers it's internal cleanup mechanism regardless
> of whether getIfPresent gets a hit has been implemented in
> [https://github.com/ben-manes/caffeine/commit/7239bb0dda2af1e7301e8f66a5df28215b5173bc]
> and is due to be released in caffeine 2.8.4
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org
