Skip to content
View in the app

A better way to browse. Learn more.

Unraid

A full-screen app on your home screen with push notifications, badges and more.

To install this app on iOS and iPadOS
  1. Tap the Share icon in Safari
  2. Scroll the menu and tap Add to Home Screen.
  3. Tap Add in the top-right corner.
To install this app on Android
  1. Tap the 3-dot menu (⋮) in the top-right corner of the browser.
  2. Tap Add to Home screen or Install app.
  3. Confirm by tapping Install.

SAMBA drops connection

Featured Replies

Hello,

 

I am having a major issue with samba on unRAID 6.0 beta 6 and 7. Samba drops connection in a middle of file copying of large files 5+ GB. I've turned the log on and got the following messages:

 

STATUS=daemon 'smbd' finished starting up and ready to serve connections

WARNING: The "null passwords" option is deprecated

 

[2014/08/24 09:15:03.310930,  1] ../source3/smbd/service.c:1130(close_cnum)

  windows81pro-vm (ipv4:192.168.0.127:50350) closed connection to service home.dir

[2014/08/24 09:16:58.905444,  1] ../source3/smbd/service.c:1130(close_cnum)

  windows81pro-vm (ipv4:192.168.0.127:50354) closed connection to service home.dir

 

Any help/suggestions/fixes will be greatly appreciated.

My system is running on SuperMicro X10SLH-F motherboard with 24GB of ECC RAM and 14x Seagate ST4000DM000 (4TB) hard drives.

 

Thanks,

Michael Abadjiev

 

Please refer to the Guidelines for Defect Reports post.  You did not provide enough information for me to troubleshoot this problem - which is another way of saying: I can transfer huge files all day long and never see this issue.  There must be something unique about your config/environment.

  • Author

As requested, here is the information

 

unRAID OS Version:

unRAIDServer-6.0-beta7-x86_64

 

Description:

I am having a major issue with samba on unRAID 6.0 beta 6 and 7. Samba drops connection in a middle of file copying of large files 5+ GB, so the file is partially copied. I've turned the samba log on and got the following messages:

 

STATUS=daemon 'smbd' finished starting up and ready to serve connections

WARNING: The "null passwords" option is deprecated

 

[2014/08/24 09:15:03.310930,  1] ../source3/smbd/service.c:1130(close_cnum)

  windows81pro-vm (ipv4:192.168.0.127:50350) closed connection to service home.dir

[2014/08/24 09:16:58.905444,  1] ../source3/smbd/service.c:1130(close_cnum)

  windows81pro-vm (ipv4:192.168.0.127:50354) closed connection to service home.dir

 

How to reproduce:

Stock setup of samba

 

Expected results:

Should copy the file as in version 5.0.5

 

Actual results:

Copying is terminated unexpected.

 

Other information:

 

Works fine on unRAID 5.0.5

 

Where is the Windows 8.1 pro VM running?  Is this on the unRAID server or another system in your setup?

  • Author

Hello jonp,

 

Thanks for your instant response.

 

Windows 8.1 Pro is running on my Mac under VMWare Fusion. The samba drop happened on all of my Macs and Windows Virtual machines, I don't have physical Windows machines.

Also, I've noticed that parity check coincidently was running on the server (unRAID 6.0 beta7) so I've stopped it.

I was copying 29GB ova file and samba drops connection after 5GB of copying. I just found that 11GB file copy is successful (tried few times)

 

Thanks,

mabadjiev

  • Author

Just FYI.

 

I found that the problem is really bad if the parity check is running.

So, to recreate the problem:

1. Start parity check with error correction/fix.

2. Wait few minutes and start copying a large file (more than 10 GB)

I was able to recreate the issue multiple times now.

 

Thanks,

mabadjiev

 

  • 1 month later...

Is this still persisting for you in the current beta?  I cannot reproduce from my end.

  • Author

Thanks for your reply. ;)

unRAID 6 beta 10a is a bit better, but still happens with large files, I've tried 29GB file without parity check running and it happened. If you turn on parity check happens instantly.

Ok, to make sure we're clear:

 

1)  unRAID server booted, no plugins installed or running?

2)  From Windows VM, you connect using file explorer to //tower (or whatever your server name is)

3)  You begin a file transfer where the source of the content is inside the vdisk of the Windows VM and copy to an unRAID user share over SMB

4)  Copy proceeds like normal, but cuts out in the middle somewhere.  Cuts out faster if this is done during a parity check.

 

Correct?

A few follow up questions:

 

1)  Is this a cache enabled or cache only user share?

2)  What write speed do you witness during the file transfer?  Does it fluctuate or remain consistent?

3)  Does it ever pause in the midst of the transfer to only resume moments later?

4)  Is this a private or secure user share?

 

I'm testing a 40GB file right now to see if I can recreate in beta 10a.  I am not running a parity check, but will re-run the test after this copy completes if it doesn't cut out...

unRAID 6 beta 10a is a bit better, but still happens with large files, I've tried 29GB file without parity check running and it happened. If you turn on parity check happens instantly.

 

Please provide the syslog for this session, zipped if necessary.

  • Author

Yes, I've tried all that. I also tried from a Mac - same result.

 

I have two unRAID servers and happens on both. I even went one step further and replace the switch that I was using - same thing.

 

My Servers are:

SuperMicro X10SLH-F with 2x Supermicro AOC-SAS2LP-MV8  total 15 disks and Intel Xeon E3-1245V3 Haswell 3.4GHz 24GB RAM

and

ASRock Motherboard E3C226D2I with one SupermicroAOC-SASLP-MV8  total 8 disks and Intel Xeon E3-1230 V3 3.3GHz Quad-Core Processor 16GB RAM

 

Thanks.

  • Author

Ok, I will do that by the end of the day...

f51432429d22888933bba649b690842a.jpg

 

Doing a copy of a 46gb vdisk from erics win81 workstation... 

Well my test completed with no "drop" and now it's running again while a parity check is going.  The write speed is definitely reduced (obviously) but it's still going.  We'll keep an eye on this and see if it drops out.

  • Author

Here is the attachment and screenshot on a Mac.

syslog.tar.gz

Screen_Shot_2014-10-02_at_4_38.33_PM.png.c7a34a713f146caa0e4e3f9a9c792763.png

Ok, I'm not an expert at decyphering syslogs, so someone else may need to examine this, but a simple

 

cat syslog | grep error 

 

showed me this:

 

Sep 29 20:00:22 unraid kernel: sas: ata7: end_device-1:0: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata7: end_device-1:0: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata8: end_device-1:1: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata7: end_device-1:0: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata8: end_device-1:1: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata9: end_device-1:2: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata7: end_device-1:0: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata8: end_device-1:1: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata9: end_device-1:2: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata10: end_device-1:3: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata7: end_device-1:0: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata8: end_device-1:1: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata9: end_device-1:2: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata10: end_device-1:3: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata11: end_device-1:4: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata7: end_device-1:0: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata8: end_device-1:1: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata9: end_device-1:2: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata10: end_device-1:3: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata11: end_device-1:4: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata12: end_device-1:5: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata7: end_device-1:0: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata8: end_device-1:1: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata9: end_device-1:2: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata10: end_device-1:3: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata11: end_device-1:4: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata12: end_device-1:5: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata13: end_device-1:6: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata7: end_device-1:0: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata8: end_device-1:1: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata9: end_device-1:2: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata10: end_device-1:3: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata11: end_device-1:4: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata13: end_device-1:6: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata12: end_device-1:5: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata14: end_device-1:7: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata15: end_device-8:0: dev error handler

 

Not sure what that means exactly (might need Tom or someone else to review).  Still going to pour through those logs some more.

 

Also, there is a log file under /var/log/samba I think (maybe two).  Can you post those as well for us to examine?

Just an FYI, two big copies complete in my tests today (first was 46GB, second was over 50GB) and the second test was while a parity check was running.  No cut outs, no issues.  Lack of repeatability could mean this isn't a bug, but another problem with the system.  Not ruling out the bug possibility though until we know that's the case...

  • Author

Interesting, these errors are on September 29, 2014, however I've got the error today....

Something to be investigated I would think.

Sep 29 20:00:22 unraid kernel: sas: ata7: end_device-1:0: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata8: end_device-1:1: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata9: end_device-1:2: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata10: end_device-1:3: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata11: end_device-1:4: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata13: end_device-1:6: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata12: end_device-1:5: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata14: end_device-1:7: dev error handler

Sep 29 20:00:22 unraid kernel: sas: ata15: end_device-8:0: dev error handler

Those aren't error lines, just lines where the 'dev error handler' reports that it has completed setting up a drive.  It's a very noisy handler, with a bug in its logging the end of each setup, but completely harmless.  I saw no issues with the drive systems.

 

The only Samba stop shown in this syslog is at Sep 30 19:59:30, where all services are stopped then restarted.  It looks intentional to me, occurs almost exactly 24 hours after booting, so I don't think it's the Samba drop you were referring to.  In fact, I'm wondering if the messages earlier you posted about smbd dropping were from the Mac, not the UnRAID server.  Can you confirm where you saw them?  The only other drops I saw were SSH sessions, and they looked intentional by user. (but I could be wrong)  There is stuff here that's new to me.

 

When you have one of these Samba drops, try going to another machine and trying a Samba transfer to your UnRAID server, to check whether it's still working there.

 

[quote author=mabadjiev link=topic=34799.msg331240#msg33124

0 date=1412300127]

Interesting, these errors are on September 29, 2014, however I've got the error today....

Something to be investigated I would think.

In your syslog is the following line:

Sep 29 20:00:22 unraid kernel: rtc_cmos 00:02: setting system clock to 2014-09-30 02:59:59 UTC (1412045999)

So I suspect your timezone setting (or time offset) may not be set correctly.  Perhaps a sign reversal?  There's often confusion over whether the offset is a plus from UTC or a minus.  Try checking your time settings, reversing the sign perhaps.

  • Author

Hello,

 

The disconnect happens from any OS accessing the server via CIFS. I've tried to turn the SAMBA logging on - nothing meaningful so far, it just says - Client close connection.

 

As far as the time setting I believe they are correct, internally unRAID is probably using UTC (see attached)

 

Thanks for your time.

Screen_Shot_2014-10-02_at_10_14.22_PM.png.4876231dcab0c90585629defed6e40b0.png

Hello,

 

The disconnect happens from any OS accessing the server via CIFS. I've tried to turn the SAMBA logging on - nothing meaningful so far, it just says - Client close connection.

 

As far as the time setting I believe they are correct, internally unRAID is probably using UTC (see attached)

 

Thanks for your time.

 

This is sounding more and more like a client issue, not unRAID.  All my transfer attempts from various devices seem to work fine to my systems (multiple testing). 

  • Author

Interesting on unRAID 5.0.5 none of this happens - same equipment.

I am not quite sure what do you mean by 'client issue'.

I was able to recreate the problem even on Linux (ubuntu 14.04 LTS) while mounting share via CIFS.

So if all the clients Linux, Windows and Mac are having this issue, I would think more of a client-server issue and since that does not happen on unRAID 5.0.5 more of a server issue.

Interesting on unRAID 5.0.5 none of this happens - same equipment.

I am not quite sure what do you mean by 'client issue'.

I was able to recreate the problem even on Linux (ubuntu 14.04 LTS) while mounting share via CIFS.

So if all the clients Linux, Windows and Mac are having this issue, I would think more of a client-server issue and since that does not happen on unRAID 5.0.5 more of a server issue.

Good point. This is just really odd that I can't recreate. Need to ponder this one for a little bit.

OK, let's try this.  Do you have it so you can browse your disk shares over SMB directly?  If so, can you try an SMB copy to just a single disk for me? If that fails, stop the array, turn off user shares, then start the array and repeat that test again.

 

PS: let me know if you need more detailed instructions on how to do all of that.

Archived

This topic is now archived and is closed to further replies.

Account

Navigation

Search

Search

Configure browser push notifications

Chrome (Android)
  1. Tap the lock icon next to the address bar.
  2. Tap Permissions → Notifications.
  3. Adjust your preference.
Chrome (Desktop)
  1. Click the padlock icon in the address bar.
  2. Select Site settings.
  3. Find Notifications and adjust your preference.