kudu-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Saeid Sattari <saeid.satt...@gmail.com>
Subject Re: WAL directory is full
Date Thu, 17 May 2018 09:27:01 GMT
Hi,

Sorry for the belated response. The version I am using is 1.6. I didn't
check with cluster ksck. Since it was a bit urgent I removed the
datanode from kudu cluster and after being sure that all missing tablets
are replicated I formatted the kudu file system and re-joined that node to
the cluster. Thank you anyway.

Regards,
Saeid

On Mon, May 14, 2018 at 10:42 PM, Mike Percy <mpercy@apache.org> wrote:

> Hi Saeid,
> What version of Kudu are you running? Do you see any errors when you run
> "sudo -u kudu kudu cluster ksck" on the cluster?
>
> Mike
>
> On Fri, May 11, 2018 at 5:12 AM, Saeid Sattari <saeid.sattari@gmail.com>
> wrote:
>
>> Hi all,
>>
>> I assigned a 100GB SSD disks to WAL on each node in my cluster. Recently,
>> I realized that some nodes replicated their cfiles to other nodes due to
>> the insufficient space. I found this flag--log_max_segments_to_retain
>> <https://kudu.apache.org/docs/configuration_reference.html#kudu-master_log_max_segments_to_retain>
>> that control the number of past logs to preserve but it is currently
>> unsupported. Do you have any idea or experience on solving this problem?
>> Thank you in advance.
>>
>> Regards,
>> Saeid
>>
>
>

Mime
View raw message