nifi-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Michael Moser <moser...@gmail.com>
Subject Re: Unable to checkpoint FlowFile Repository (No space left on device)
Date Thu, 14 Dec 2017 16:27:43 GMT
Maximum number of inodes is defined when you build a file system on a
device.  You would have to backup your data, rebuild your file system and
tell it to allocate more inodes than the default (an mkfs option, I
think?), then restore your data.

You can turn off the NiFi content_repository archive, if you don't need
that feature, by setting nifi.content.repository.archive.enabled=false in
your nifi.properties.

-- Mike


On Thu, Dec 14, 2017 at 2:22 AM, 尹文才 <batman713@gmail.com> wrote:

> One Strange thing one of our testers found was that we're using the default
> 12hours archive and 50% disk space configuration, he noticed that when nifi
> removed the archived files inside the content_repository, he checked the
> inodes ' count didn't go down, then he tried to remove some of the archived
> files inside using the rm command, the inodes's count did go down.
>
> /Ben
>
> 2017-12-14 9:49 GMT+08:00 尹文才 <batman713@gmail.com>:
>
> > Hi Michael, the no space left on device occurred again and I checked the
> > inodes at the time and found it was indeed full, why would the inodes
> > become full and are there any solutions to get around this problem?
> Thanks.
> >
> > /Ben
> >
> > 2017-12-13 13:36 GMT+08:00 尹文才 <batman713@gmail.com>:
> >
> >> Hi Michael, I checked the system available inodes by running df -i
> >> command and there're quite enough inodes in the system. I then removed
> all
> >> the files in all repository folders and restarted
> >> the system, I couldn't see the error again. I will continue to track the
> >> problem to see what's causing it, but it seems not relevant to the inode
> >> use-up reason you mentioned. Thanks.
> >>
> >> /Ben
> >>
> >> 2017-12-12 23:45 GMT+08:00 Michael Moser <moser.mw@gmail.com>:
> >>
> >>> Greetings Ben,
> >>>
> >>> The "No space left on device" error can also be caused by running out
> of
> >>> inodes on your device.  You can check this with "df -i".
> >>>
> >>> -- Mike
> >>>
> >>>
> >>> On Tue, Dec 12, 2017 at 1:36 AM, 尹文才 <batman713@gmail.com> wrote:
> >>>
> >>> > sorry that I forgot to mention the environment that caused this
> >>> problem,
> >>> > I'm using the latest nifi 1.4.0 release and installed it on centos
7.
> >>> >
> >>> > 2017-12-12 14:35 GMT+08:00 尹文才 <batman713@gmail.com>:
> >>> >
> >>> > > Hi guys, I'm running into a very weird problem, I wrote a processor
> >>> > > specifically to extract some data
> >>> > > and I found starting from yesterday it kept showing errors in
the
> >>> log, as
> >>> > > below:
> >>> > >
> >>> > > 2017-12-12 14:01:04,661 INFO [pool-10-thread-1] o.a.n.c.r.
> >>> > WriteAheadFlowFileRepository
> >>> > > Initiating checkpoint of FlowFile Repository
> >>> > > 2017-12-12 14:01:04,676 ERROR [pool-10-thread-1] o.a.n.c.r.
> >>> > WriteAheadFlowFileRepository
> >>> > > Unable to checkpoint FlowFile Repository due to
> >>> > > java.io.FileNotFoundException: ../flowfile_repository/
> >>> > partition-5/96.journal
> >>> > > (No space left on device)
> >>> > > java.io.FileNotFoundException: ../flowfile_repository/
> >>> > partition-5/96.journal
> >>> > > (No space left on device)
> >>> > >         at java.io.FileOutputStream.open0(Native Method)
> >>> > >         at java.io.FileOutputStream.open(
> FileOutputStream.java:270)
> >>> > >         at java.io.FileOutputStream.<init
> >>> >(FileOutputStream.java:213)
> >>> > >         at java.io.FileOutputStream.<init
> >>> >(FileOutputStream.java:162)
> >>> > >         at org.wali.MinimalLockingWriteAheadLog$
> Partition.rollover(
> >>> > > MinimalLockingWriteAheadLog.java:779)
> >>> > >         at org.wali.MinimalLockingWriteAheadLog.checkpoint(
> >>> > > MinimalLockingWriteAheadLog.java:528)
> >>> > >         at org.apache.nifi.controller.repository.
> >>> > > WriteAheadFlowFileRepository.checkpoint(WriteAheadFlowFileRe
> >>> pository.
> >>> > > java:451)
> >>> > >         at org.apache.nifi.controller.repository.
> >>> > > WriteAheadFlowFileRepository$1.run(WriteAheadFlowFileRepository.
> >>> > java:423)
> >>> > >         at java.util.concurrent.Executors$RunnableAdapter.
> >>> > > call(Executors.java:511)
> >>> > >         at java.util.concurrent.FutureTask.runAndReset(
> >>> > > FutureTask.java:308)
> >>> > >         at java.util.concurrent.ScheduledThreadPoolExecutor$
> >>> > > ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.
> java:180)
> >>> > >         at java.util.concurrent.ScheduledThreadPoolExecutor$
> >>> > > ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:294)
> >>> > >         at java.util.concurrent.ThreadPoolExecutor.runWorker(
> >>> > > ThreadPoolExecutor.java:1142)
> >>> > >         at java.util.concurrent.ThreadPoolExecutor$Worker.run(
> >>> > > ThreadPoolExecutor.java:617)
> >>> > >         at java.lang.Thread.run(Thread.java:745)
> >>> > >
> >>> > >
> >>> > > I noticed the log mentioned no space left on device and I went
to
> >>> check
> >>> > > the available space and found 33G left. Does anyone know what
could
> >>> > > possibly cause this and how to resolve this problem, thanks
> >>> > >
> >>> > > /Ben
> >>> > >
> >>> >
> >>>
> >>
> >>
> >
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message