Power issue caused crash - now volume will not mount

[Louis van Dyk]

Hi

I am getting the below error after my power utility had some nasty up and downs for about a day.  I switched the server off once I discovered it - but by the even the UPS had run out of juice.

Nevertheless, once I eventually tried to reboot, I get this on mounting the SDFS volume.  I'm running on Fedora 28 (let's call it Red Hat for the sake of compliance).  I've tried renaming the hdb folder and adding rebuild-hashtable="true" to the <volumes> tag in the xml file.  I still got the error.

Please help to get the volume mounted again.

Running Program SDFS Version 3.7.7.7 build date 2018-08-08 02:33
reading config file = /etc/sdfs/pool0-volume-cfg.xml
Unable to initiate ChunkStore
java.io.IOException: java.io.EOFException
    at org.opendedup.sdfs.filestore.BatchFileChunkStore.init(BatchFileChunkStore.java:275)
    at org.opendedup.sdfs.servers.HashChunkService.<init>(HashChunkService.java:66)
    at org.opendedup.sdfs.servers.HCServiceProxy.init(HCServiceProxy.java:154)
    at org.opendedup.sdfs.servers.SDFSService.start(SDFSService.java:86)
    at fuse.SDFS.MountSDFS.setup(MountSDFS.java:214)
    at fuse.SDFS.MountSDFS.init(MountSDFS.java:253)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:498)
    at org.apache.commons.daemon.support.DaemonLoader.load(DaemonLoader.java:207)
Caused by: java.io.EOFException
    at java.io.ObjectInputStream$PeekInputStream.readFully(ObjectInputStream.java:2335)
    at java.io.ObjectInputStream$BlockDataInputStream.readShort(ObjectInputStream.java:2804)
    at java.io.ObjectInputStream.readStreamHeader(ObjectInputStream.java:802)
    at java.io.ObjectInputStream.<init>(ObjectInputStream.java:299)
    at org.opendedup.sdfs.filestore.BatchFileChunkStore.init(BatchFileChunkStore.java:253)
    ... 10 more
Service exit with a return value of 255

[Louis van Dyk]

Anyone able to help me?

I still cannot mount the sdfs filesystem.

[Sam Silverberg]

Hi Louis,

I would be happy to help.

Can you send the log output so I can review.

--
You received this message because you are subscribed to the Google Groups "dedupfilesystem-sdfs-user-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dedupfilesystem-sdfs-u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Louis van Dyk]

Hi Sam

THANK you!

I've just tried to mount it again, and then tarred the whole log folder. See attached.

Regards,
Louis

To unsubscribe from this group and stop receiving emails from it, send an email to dedupfilesystem-sdfs-user-discuss+unsubscribe@googlegroups.com.

[Louis van Dyk]

Sam ?

[Louis van Dyk]

Please please please please please please please can somebody help me?

[Sam Silverberg]

I am looking at this right now.

On Thu, Nov 1, 2018 at 5:05 PM Louis van Dyk <louis....@gmail.com> wrote:

Please please please please please please please can somebody help me?

--

You received this message because you are subscribed to the Google Groups "dedupfilesystem-sdfs-user-discuss" group.

To unsubscribe from this group and stop receiving emails from it, send an email to dedupfilesystem-sdfs-u...@googlegroups.com.

[Louis van Dyk]

Hi Sam .... any news?

On Friday, 2 November 2018 17:10:20 UTC+2, Sam Silverberg wrote:

I am looking at this right now.

On Thu, Nov 1, 2018 at 5:05 PM Louis van Dyk <louis....@gmail.com> wrote:

Please please please please please please please can somebody help me?

--
You received this message because you are subscribed to the Google Groups "dedupfilesystem-sdfs-user-discuss" group.

To unsubscribe from this group and stop receiving emails from it, send an email to dedupfilesystem-sdfs-user-discuss+unsubscribe@googlegroups.com.

[Louis van Dyk]

Hi Sam  ....   do you have any advice for me yet?

[Sam Silverberg]

Louis - It looks as if there are some corrupt archives on the disk. Are there any file is /opt/sdfs/volumes/volume-name/chunkstore/chunks/outgoing ?

On Sat, Nov 17, 2018 at 1:44 PM Louis van Dyk <louis....@gmail.com> wrote:

Hi Sam  ....   do you have any advice for me yet?

On Friday, 9 November 2018 04:59:46 UTC+2, Louis van Dyk wrote:

Hi Sam .... any news?

On Friday, 2 November 2018 17:10:20 UTC+2, Sam Silverberg wrote:

I am looking at this right now.

On Thu, Nov 1, 2018 at 5:05 PM Louis van Dyk <louis....@gmail.com> wrote:

Please please please please please please please can somebody help me?

--
You received this message because you are subscribed to the Google Groups "dedupfilesystem-sdfs-user-discuss" group.

To unsubscribe from this group and stop receiving emails from it, send an email to dedupfilesystem-sdfs-u...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

You received this message because you are subscribed to the Google Groups "dedupfilesystem-sdfs-user-discuss" group.

To unsubscribe from this group and stop receiving emails from it, send an email to dedupfilesystem-sdfs-u...@googlegroups.com.

[Louis van Dyk]

Hi Sam.  No there is no folder called "outgoing" in chunkstore/chunks.

Anything else I can check?

[Louis van Dyk]

[Louis van Dyk]

Dear group members and Sam

<BEGGING MODE ON>

I have not had access to my SDFS file system for 2 months now.  Do I need to kiss that data goodbye?  Is there no way to remount it?  Please could someone help me get this volume mounted again?

<BEGGING MODE OFF>

Many thanks

Louis

[Sam Silverberg]

Louis - It looks like the data is corrupt. I am sorry that I have not gotten back to you and if you would like we could setup a webex an see if I can fix it. Let me know it that works for you.

[Louis van Dyk]

Hi Sam.  Thanks for coming back to me.  I was using this as my backup store.  Since it is corrupt I will trash it and restart.  I don't want to waste your time further on it as it's not business critical.  How often does this sort of thing happen though?  I would hate to be depending on OpenDeDup to store my more critical data only to have something like this happen to it.