phoenix-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Vlad Krava (Jira)" <>
Subject [jira] [Updated] (PHOENIX-5250) The accumulated wal files can't be cleaned
Date Fri, 25 Oct 2019 21:29:00 GMT


Vlad Krava updated PHOENIX-5250:
    Comment: was deleted

(was: [~elserj] [~jaanai]

I assume this issue appears when IndexedWALEditCodec is enabled (must have for LOCAL and
GLOBAL index) as it uses IndexedKeyValue, so it's not only related to the filtering after GLOBAL
index sync.

Is there any workaround you may propose besides RS restart, taking into consideration that
for my needs SECONDARY index is must have?)

> The accumulated wal files can't be cleaned
> ------------------------------------------
>                 Key: PHOENIX-5250
>                 URL:
>             Project: Phoenix
>          Issue Type: Bug
>            Reporter: Jaanai Zhang
>            Priority: Blocker
>             Fix For: 5.1.0
>         Attachments: image-2019-04-19-21-31-27-888.png
> Because of the modification of HBASE-20781,  the faked WALEdits won't be filtered,
all WALEdits will be saved into Memstore with a status that inMemstore is true(uses WAL->append
> !image-2019-04-19-21-31-27-888.png|width=755,height=310!
> The family array of IndexedKeyValue is WALEdit.METAFAMILY that is used to describe
a fake WALEdit, and it will put into Memstore with WALedits of data table during sync global
> WAL files can't be cleaned, except for restarting RS, Many WAL files will decrease the
percent of disk free.  
> !|width=422,height=159!

This message was sent by Atlassian Jira

View raw message