.
-Original Message-
From: Sean Busbey
Sent: Tuesday, January 26, 2021 5:46 PM
To: Hbase-User
Subject: Re: Region server idle
EXTERNAL
You can change the logging for just the balancer system so that not so much
goes into the logs. The logger to change
.
-Original Message-
From: Sean Busbey
Sent: Tuesday, January 26, 2021 5:46 PM
To: Hbase-User
Subject: Re: Region server idle
EXTERNAL
You can change the logging for just the balancer system so that not so much
goes into the logs. The logger to change is
org.apache.hadoop.hbase.master.balancer
; Sent: Tuesday, January 26, 2021 3:59 PM
> To: Hbase-User
> Subject: Re: Region server idle
>
> EXTERNAL
>
> Hi Marc!
>
> Did turning up the balancer's logging confirm that it is actually making
> plans? Did it confirm what weights are being used for the various balancer
.
-Original Message-
From: Sean Busbey
Sent: Tuesday, January 26, 2021 3:59 PM
To: Hbase-User
Subject: Re: Region server idle
EXTERNAL
Hi Marc!
Did turning up the balancer's logging confirm that it is actually making plans?
Did it confirm what weights are being used for the various balancer
were well-integrated within a day or so.
>
> -Original Message-
> From: Josh Elser
> Sent: Sunday, January 24, 2021 10:22 PM
> To: user@hbase.apache.org
> Subject: Re: Region server idle
>
> EXTERNAL
>
> Yes, each RegionServer has its own writeahead log (wh
in
October 2020 and they were well-integrated within a day or so.
-Original Message-
From: Josh Elser
Sent: Sunday, January 24, 2021 10:22 PM
To: user@hbase.apache.org
Subject: Re: Region server idle
EXTERNAL
Yes, each RegionServer has its own writeahead log (which is named with that RS
sage-
From: Josh Elser
Sent: Tuesday, January 12, 2021 4:56 PM
To: user@hbase.apache.org
Subject: Re: Region server idle
EXTERNAL
Yes, in general, HDFS rebalancing will cause a decrease in the performance of
HBase as it removes the ability for HBase to short-circuit some read logic. It
sh
gt; -Original Message-
> From: Josh Elser
> Sent: Monday, January 11, 2021 5:34 PM
> To: user@hbase.apache.org
> Subject: Re: Region server idle
>
> EXTERNAL
>
> The Master stacktrace you have there does read as a bug, but it shouldn't be
> affecting balancing.
&
, 2021 5:34 PM
To: user@hbase.apache.org
Subject: Re: Region server idle
EXTERNAL
The Master stacktrace you have there does read as a bug, but it shouldn't be
affecting balancing.
That Chore is doing work to apply space quotas, but your quota here is only
doing RPC (throttle) quotas. Might
compaction.
Once again, experience in this area may be giv8ing me misinformation.
-Original Message-
From: Josh Elser
Sent: Monday, January 11, 2021 5:34 PM
To: user@hbase.apache.org
Subject: Re: Region server idle
EXTERNAL
The Master stacktrace you have there does read as a bug
pins
Sent: Monday, January 11, 2021 9:52 AM
To: user@hbase.apache.org
Subject: RE: Region server idle
EXTERNAL
I tried. Appears to have failed reading data from hbase:meta. These are
repeated errors for the whole run of list_quotas.
A balance task was run on Friday. It took 9+ hours. The a
used by GSSException:
No valid credentia
ls provided (Mechanism level: Failed to find any Kerberos tgt)]]
-Original Message-
From: Stack
Sent: Saturday, January 9, 2021 1:52 AM
To: Hbase-User
Subject: Re: Region server idle
E
on:
No valid credentia
ls provided (Mechanism level: Failed to find any Kerberos tgt)]]
-Original Message-
From: Stack
Sent: Saturday, January 9, 2021 1:52 AM
To: Hbase-User
Subject: Re: Region server idle
EXTERNAL
Looking at code around except
lang.Thread.run(Thread.java:748)
>
> -Original Message-
> From: Marc Hoppins
> Sent: Friday, January 8, 2021 10:57 AM
> To: user@hbase.apache.org
> Subject: RE: Region server idle
>
> EXTERNAL
>
> So, I tried decommission that RS and recommission it. No change.
-
From: Marc Hoppins
Sent: Friday, January 8, 2021 10:57 AM
To: user@hbase.apache.org
Subject: RE: Region server idle
EXTERNAL
So, I tried decommission that RS and recommission it. No change. Server still
idle.
Tried decommission another server and see if HBASE sets itself right. Now I
have two
-Original Message-
From: Sean Busbey
Sent: Thursday, January 7, 2021 7:30 PM
To: Hbase-User
Subject: Re: Region server idle
EXTERNAL
Sounds like https://issues.apache.org/jira/browse/HBASE-24139
The description of that jira has a workaround.
On Thu, Jan 7, 2021, 05:23 Marc Hoppins wrote
Sounds like https://issues.apache.org/jira/browse/HBASE-24139
The description of that jira has a workaround.
On Thu, Jan 7, 2021, 05:23 Marc Hoppins wrote:
> Hi all,
>
> I have a setup with 67 region servers. 29 Dec one system had to be shut
> down to have EMM module swapped out - which took
out that MEMSTORE and STOREFILES are also 0, and
> that USED HEAP is only <200MB, whereas all other hosts heap used is >16GB
> and <23GB.
>
> -Original Message-
> From: Marc Hoppins
> Sent: Thursday, January 7, 2021 12:23 PM
> To: user@hbase.apache.org
> Subject: Region
t: Region server idle
EXTERNAL
Hi all,
I have a setup with 67 region servers. 29 Dec one system had to be shut down to
have EMM module swapped out - which took one work day. Host was back online 30
Dec.
My HBASE is very basic so I appreciate your patience.
My understanding of the defaults that
Hi all,
I have a setup with 67 region servers. 29 Dec one system had to be shut down to
have EMM module swapped out - which took one work day. Host was back online 30
Dec.
My HBASE is very basic so I appreciate your patience.
My understanding of the defaults that are setup is that a major
20 matches
Mail list logo
