Error in stream protocol: End of stream

[Eamonn Sullivan]

Binary package hint: sbackup

2010-10-16 15:06:53,103 - ERROR: An error occurred during the backup: Error in stream protocol: End of stream
2010-10-16 15:06:53,241 - ERROR: Uncaught exception: The specified location is not mounted
2010-10-16 15:06:53,243 - ERROR: Traceback (most recent call last):
  File "/usr/bin/sbackup", line 42, in <module>
    retc = main(sys.argv)
  File "/usr/share/sbackup/sbackup/backupproc.py", line 586, in main
    exitcode = sbackup_app.run()
  File "/usr/share/sbackup/sbackup/backupproc.py", line 566, in run
    self.__exitcode = self.__backupproc.run()
  File "/usr/share/sbackup/sbackup/backupproc.py", line 288, in run
    self.__on_backup_error(error)
  File "/usr/share/sbackup/sbackup/backupproc.py", line 311, in __on_backup_error
    self.__exitcode = self.__bprofilehdl.finish(error)
  File "/usr/share/sbackup/sbackup/core/profile_handler.py", line 315, in finish
    self.__copylogfile()
  File "/usr/share/sbackup/sbackup/core/profile_handler.py", line 286, in __copylogfile
    _op.copyfile(logf_src, logf_target)
  File "/usr/share/sbackup/sbackup/fs_backend/_gio_utils.py", line 472, in copyfile
    _src, _dest = cls._prepare_copy(_src, _dest)
  File "/usr/share/sbackup/sbackup/fs_backend/_gio_utils.py", line 500, in _prepare_copy
    if cls.__isdir(dst_gfile):
  File "/usr/share/sbackup/sbackup/fs_backend/_gio_utils.py", line 651, in __isdir
    _ftype = cls._query_file_type(gfile)
  File "/usr/share/sbackup/sbackup/fs_backend/_gio_utils.py", line 670, in _query_file_type
    cancellable = None)
Error: The specified location is not mounted

[Eamonn Sullivan (eamonn-sullivan)]

Please tell me what further information you need. I have four Ubuntu computers running in the house, all running 10.10. One of them, a netbook, gets this error when I try to run sbackup. It has an identical configuration as the other three. It's try to backup to ssh://root@<mybackupserver.local>/var/backup/. It hasn't succeeded once yet. I'm trying to do the first full backup. I tested the connection to the remote server and get a green checkmark. I also can do "ssh root@<mybackupserver.local> whoami" successfully from the netbook.

[Eamonn Sullivan (eamonn-sullivan)]

Attaching the log file from my latest attempt. I've also tried to remove and purge sbackup and reinstall and I've installed the optional fuse add-on (sbackup-fuse or something or other), even that's not installed on the three other computers. The other difference with the netbook is that it is connected to the network over wireless, while the other three are wired, if that helps.

[Eamonn Sullivan (eamonn-sullivan)]

I put a try/except block around the line that's failing in _gio_utils.py and it's failing when trying to write excludes.list, if that's any help. The path (ssh://...) and the data look fine to me, but the writer of this will probably know more.

This is what the target directory looks like at the point of failure, so clearly it *is* succeeding at writing some of the files (and you can see that the open of excludes.list appears to have worked):

-rw-r--r-- 1 root root 213 2010-10-16 17:51 excludes
-rw-r--r-- 1 root root 0 2010-10-16 17:51 excludes.list
-rw-r--r-- 1 root root 6 2010-10-16 17:51 format
-rw-r--r-- 1 root root 10 2010-10-16 17:51 includes.list
-rw-r--r-- 1 root root 58204 2010-10-16 17:51 packages

[Eamonn Sullivan (eamonn-sullivan)]

Found a workaround: If I remove the [cC]ache regex in the exclude list, it works. My kids are
heavy browser users (mostly chrome, but also firefox). The exclude.list is mostly lots of lines of the form:

/home/<user>/.cache/google-chrome/...

I see about 8,000 lines of it. This workaround means I'm backing up some 8,000 unnecessary files, but it'll do for now.

[Nech (gerard-guadall)]

Yes Eamoon Sullivan, very good solution. I put less files and now I was well! Thank you very much.

[Jean-Peer Lorenz (peer.loz)]

Thank you for choosing SBackup and for taking the time reporting this bug.

In order to track down the cause of the problem some more information are required. Please answer the following questions:

* what distribution do you use?
* what version of sbackup is installed; from where was it installed (PPA, from source)?
* Is the bug reproducible; does it happen every time?
* do you run sbackup as superuser (root) or as regular user?
* are you able to open your backup destination using Nautilus/command line?
* Is this issue still valid?

At first glance, your specific error (Input/output error) looks like an underlying problem to me (e.g. network connection) not directly caused by sbackup.

Thanks for your help.

[Eamonn Sullivan (eamonn-sullivan)]

> What distribution do you use

Ubuntu 10.10

> what version of sbackup

0.11.3ubuntu1. Installed from the standard repositories.

> reproducible?

yes. Happens every time.

> do you run sbackup as superuser (root)

yes.

> are you able to open your backup destination using Nautilus/command line?

Yes.

> Is this issue still valid?

Yes,

> At first glance, your specific error (Input/output error) looks like an underlying problem to me (e.g. network connection) not directly caused by sbackup.

I think you haven't read the report carefully enough. :) Sbackup works on every other computer in the house (same network) and even on the computer which is having the problem -- if I make sure the exclude file doesn't get too big. It looks like it's either run into some size limit or is timing out when building the exclude file. I think the input/output error looks to me (I'm a *nix programmer for a living) like a red herring. But, as I said earlier, you know more about it than I do.

[Jean-Peer Lorenz (peer.loz)]

Thanks for your response.

> Sbackup works on every other computer in the house (same network) and even on the
> computer which is having the problem -- if I make sure the exclude file doesn't get too big
And this leads to my *assumption* that something different goes wrong. sbackup uses the same code for copying/writing of small, medium and large files :-) So, this shouldn't make any difference. Difficult to track down.

Anyway, some more questions:
* do you have issues when you limit the size of your excludes file?
* did you ever had problems when using nautilus etc.?

I've found a similar bug related to GIO/GVFS (sbackup uses it too) in LP (https://bugs.launchpad.net/ubuntu/+source/gvfs/+bug/222043) and Bugzilla (https://bugzilla.gnome.org/show_bug.cgi?id=545229).

> The other difference with the netbook is that it is connected to the network over wireless
Interesting, have a look here: http://ubuntuforums.org/showthread.php?t=1278655

Thanks again for your help.

[Eamonn Sullivan (eamonn-sullivan)]

> Do you have issues when you limit the size of your excludes file?

No, the backup worked. About 66GB of data was transferred (over the same wireless network, taking some hours).

> did you ever had problems when using nautilus etc.?

Not that I remember.

The other bugs you mention are interesting. I'll try to eliminate wireless interference: I recently changed my wireless router, but since I'm no longer using sbackup (because of this issue), I'll need to re-install and try this.

Thanks for your attention and help!

[Andrew Collier (collierab)]

hi,

has this issue been resolved? i am running sbackup-0.11.3 on ubuntu 10.10. i have excluded .gvfs as suggested above and i am using a wired network connection. so the solutions that worked for eamonn do not seem to work for me.

the end of the log gives:

2011-02-18 08:27:42,402 - INFO: Summary of backup
2011-02-18 08:27:42,403 - INFO: Number of directories: 6416.
2011-02-18 08:27:42,403 - INFO: Total number of files: 105138.
2011-02-18 08:27:42,404 - INFO: Number of symlinks: 367.
2011-02-18 08:27:42,405 - INFO: Number of files included in snapshot: 105138.
2011-02-18 08:27:42,405 - INFO: Number of new files (also included): 0.
2011-02-18 08:27:42,406 - INFO: Number of files skipped in incremental snapshot: 0.
2011-02-18 08:27:42,406 - INFO: Number of items forced to be excluded: 6238.
2011-02-18 08:27:42,407 - INFO: Number of items to be excluded by config: 5519.
2011-02-18 08:27:42,408 - INFO: Maximum free size required is '91097 MiB 906 KiB 229'.
2011-02-18 08:27:42,408 - WARNING: Unable to query available space on target: Operation not supported
2011-02-18 08:27:42,408 - INFO: Snapshot is being committed
2011-02-18 08:27:50,843 - ERROR: An error occurred during the backup: Error in stream protocol: End of stream
2011-02-18 08:27:50,978 - ERROR: Uncaught exception: The specified location is not mounted
2011-02-18 08:27:50,979 - ERROR: Traceback (most recent call last):
  File "/usr/bin/sbackup", line 42, in <module>
    retc = main(sys.argv)
  File "/usr/share/sbackup/sbackup/backupproc.py", line 586, in main
    exitcode = sbackup_app.run()
  File "/usr/share/sbackup/sbackup/backupproc.py", line 566, in run
    self.__exitcode = self.__backupproc.run()
  File "/usr/share/sbackup/sbackup/backupproc.py", line 288, in run
    self.__on_backup_error(error)
  File "/usr/share/sbackup/sbackup/backupproc.py", line 311, in __on_backup_error
    self.__exitcode = self.__bprofilehdl.finish(error)
  File "/usr/share/sbackup/sbackup/core/profile_handler.py", line 315, in finish
    self.__copylogfile()
  File "/usr/share/sbackup/sbackup/core/profile_handler.py", line 286, in __copylogfile
    _op.copyfile(logf_src, logf_target)
  File "/usr/share/sbackup/sbackup/fs_backend/_gio_utils.py", line 472, in copyfile
    _src, _dest = cls._prepare_copy(_src, _dest)
  File "/usr/share/sbackup/sbackup/fs_backend/_gio_utils.py", line 500, in _prepare_copy
    if cls.__isdir(dst_gfile):
  File "/usr/share/sbackup/sbackup/fs_backend/_gio_utils.py", line 651, in __isdir
    _ftype = cls._query_file_type(gfile)
  File "/usr/share/sbackup/sbackup/fs_backend/_gio_utils.py", line 670, in _query_file_type
    cancellable = None)
Error: The specified location is not mounted

i have tried to do the backup a number of times (and from different network links) and the result is always the same. has this issue maybe been fixed in the newer release (0.11.4)? please let me know if i can provide any additional information.

best regards,
andrew.

[Andrew Collier (collierab)]

here is the full log file.

[taka k. (scar)]

no, it hasn't been fixed in 0.11.4 yet. i just checked today. see bug #675644. what i have been doing as a workaround is saving the backups to another partition and then using rsync to transfer the backups to where i want them.......

[cuby (cuby)]

Happens the same thing to me in the same context.
In my case I have an enormous quantity of jar files that I want to exclude.
I'm on version 0.11.4 on Ubuntu 12.04 LTS 64 bits up to date.

[cuby (cuby)]

The destination is accessed using sftp
I get this in the end of the log:

2013-01-03 14:21:08,796 - INFO: Summary of backup
2013-01-03 14:21:08,796 - INFO: Number of directories: 21974.
2013-01-03 14:21:08,796 - INFO: Total number of files: 56033.
2013-01-03 14:21:08,796 - INFO: Number of symlinks: 1071.
2013-01-03 14:21:08,796 - INFO: Number of files included in snapshot: 56033.
2013-01-03 14:21:08,796 - INFO: Number of new files (also included): 0.
2013-01-03 14:21:08,796 - INFO: Number of files skipped in incremental snapshot: 0.
2013-01-03 14:21:08,796 - INFO: Number of items forced to be excluded: 10.
2013-01-03 14:21:08,796 - INFO: Number of items to be excluded by config: 5335.
2013-01-03 14:21:08,796 - INFO: Maximum free size required is '2973 MiB 905 KiB 957'.
2013-01-03 14:21:08,796 - WARNING: Unable to query available space on target: Operation not supported
2013-01-03 14:21:08,796 - INFO: Snapshot is being committed
2013-01-03 14:21:10,651 - ERROR: An error occurred during the backup: Error in stream protocol: End of stream
2013-01-03 14:21:10,659 - ERROR: Uncaught exception: The specified location is not mounted
2013-01-03 14:21:10,659 - ERROR: Traceback (most recent call last):
  File "/usr/bin/sbackup", line 42, in <module>
    retc = main(sys.argv)
  File "/usr/share/sbackup/sbackup/backupproc.py", line 581, in main
    exitcode = sbackup_app.run()
  File "/usr/share/sbackup/sbackup/backupproc.py", line 561, in run
    self.__exitcode = self.__backupproc.run()
  File "/usr/share/sbackup/sbackup/backupproc.py", line 290, in run
    self.__on_backup_error(error)
  File "/usr/share/sbackup/sbackup/backupproc.py", line 314, in __on_backup_error
    self.__exitcode = self.__bprofilehdl.finish(error)
  File "/usr/share/sbackup/sbackup/core/profile_handler.py", line 315, in finish
    self.__copylogfile()
  File "/usr/share/sbackup/sbackup/core/profile_handler.py", line 286, in __copylogfile
    _op.copyfile(logf_src, logf_target)
  File "/usr/share/sbackup/sbackup/fs_backend/_gio_utils.py", line 487, in copyfile
    _src, _dest = cls._prepare_copy(_src, _dest)
  File "/usr/share/sbackup/sbackup/fs_backend/_gio_utils.py", line 513, in _prepare_copy
    if cls.__isdir(dst_gfile):
  File "/usr/share/sbackup/sbackup/fs_backend/_gio_utils.py", line 664, in __isdir
    _ftype = cls._query_file_type(gfile)
  File "/usr/share/sbackup/sbackup/fs_backend/_gio_utils.py", line 683, in _query_file_type
    cancellable = None)
Error: The specified location is not mounted

[Jean-Peer Lorenz (peer.loz)]

I'm setting the status of this report to confirmed since it affects multiple users.

Unfortunately, I was not able to reproduce the error so far.

@cuby: Would you mind to post your full log file (debug mode) please?

I start to think it is related to some timing issue...help in tracking this down is *highly* appreciated. Thanks.