Bug 181780

This you can see only in error_log or also in printers.conf?
If second, attach yast logs please.
Changing serverity to Major
( Critical - Crash, loss of data, corruption of data, severe memory leak,
Major - Major loss of function )

here is some more of error log:

I [05/Jun/2006:13:17:50 -0400] Setting y2test device-uri to "smb://fordna1/jrobiso2%3aCallan04%40sic000
02.fsic.ford.com/PR32579" (was "file:/dev/null".)
I [05/Jun/2006:13:17:50 -0400] Setting y2test printer-is-accepting-jobs to 1 (was 0.)
I [05/Jun/2006:13:17:50 -0400] Setting y2test printer-state to 3 (was 5.)
I [05/Jun/2006:13:17:50 -0400] Saving printers.conf...
I [05/Jun/2006:13:17:50 -0400] New printer 'y2test' added by 'root'.
I [05/Jun/2006:13:17:50 -0400] Adding start banner page "none" to job 1.
I [05/Jun/2006:13:17:50 -0400] Adding end banner page "none" to job 1.
I [05/Jun/2006:13:17:50 -0400] Job 1 queued on 'y2test' by 'root'.
I [05/Jun/2006:13:17:50 -0400] Started filter /usr/lib/cups/filter/pstops (PID 24202) for job 1.
I [05/Jun/2006:13:17:50 -0400] Started backend /usr/lib/cups/backend/smb (PID 24203) for job 1.
E [05/Jun/2006:13:17:53 -0400] [Job 1] Unable to connect to CIFS host, will retry in 60 seconds...
E [05/Jun/2006:13:18:55 -0400] [Job 1] Unable to connect to CIFS host, will retry in 60 seconds...
E [05/Jun/2006:13:19:59 -0400] [Job 1] Unable to connect to CIFS host, will retry in 60 seconds...

Which makes sense since the url it's trying to reach now has crap in it.
I will attach my y2log. It's long . . . jump to the bottom and work backwords to find printer stuff.

Created attachment 87414 [details]
y2log (modified to remove passwords)

Regarding severity - Loss of printing ability does not constitute major loss of functionality?

I'm not trying to be snotty - just explaining my reasoning for putting it as critical.

On the first glance it looks like an encoding (locale) problem
(e.g. UTF-8 stuff instead of plain ASCII), but '3a' is the
hex code for the ':' character and '40' is the hex code for
the '@' character which are both plain ASCII.

I have no idea what (in YaST or in CUPS or elsewhere) replaces
plain ASCII characters with their hex codes.

Jonathon,
does your "MYPASSWORD" contain non-ASCII characters
(i.e. language specific characters like umlauts ...)?

I tried to reproduce it:

I enter in YaST (via [Add] -> "Network printer" -> "Print via SMB
Network Server") the following values for SMB share access:
Workgroup: SmbWorkgroup
Hostname: SmbHost
Queuename: SmbShare
User: SmbUser
Password: SmbPassword

This results in /etc/cups/printers.conf
DeviceURI smb://SmbWorkgroup/SmbUser:SmbPassword@SmbHost/SmbShare
which looks o.k. for me.

When I change this (e.g. leave Workgroup empty) I get
DeviceURI smb://SmbUser:SmbPassword@SmbHost/SmbShare
which looks also o.k. for me.

In /var/log/cups/error_log there is
---------------------------------------------------------------------
Setting smbtest device-uri to
 "smb://SmbWorkgroup/SmbUser:SmbPassword@SmbHost/SmbShare"
 (was "file:/dev/null".)
...
Setting smbtest device-uri to "smb://SmbHost/SmbShare"
 (was "smb://SmbWorkgroup/SmbUser:SmbPassword@SmbHost/SmbShare".)
---------------------------------------------------------------------
Note that I don't have such a SMB share and it is actually
not needed - one can simply enter arbitrary values for a test.

This works for me using Suse Linux 10.1 in x86_64 when I run YaST
as root with "en_GB.iso885915" locale and even when I use
the "de_DE.utf8" locale for root.

Summary:
I cannot reproduce it.

I only get this error after multiple alterations to the Connection part of the printer setup. Make a change, test it, make a change, test it, etc.  I got to wondering why it would fail on connections which I knew were perfect, until I checked into the conf it was making and the error logs. Suddenly I could see why.

My password does not contain anything wierd. Capital letters, small letters, and numbers. That's it. 8 Characters.

What is your locale settings?

Now I think it's not locale problem.
So if you can write how exactly I can reproduce it, it helps a lot.
It's hard to reproduce like: Make a change, test it, make a change, test it, etc.
And also check your /etc/cups/printers.conf to be sure this string is already corrupted there, please.

I just kept changing the user and password in the smb printer setup, then clicking "Test" from the printer overview screen. I have 6 users on the Windows box, so I was testing all six (all have "normal" passwords and usernames). After making one or two changes, it would insert those hex strings instead of the correct : and @.

If you are unable to reproduce this then we'll have to drop it. I don't want you wasting too much effort if I am the only one experiencing this.

From /etc/sysconfig/language: RC_LANG="en_US.UTF-8"

"env | grep locale" gets nothing for root, and as a user I get:
G_FILENAME_ENCODING=@locale,UTF-8,ISO-8859-15,CP1252

>If you are unable to reproduce this then we'll have to drop it. I don't want
you wasting too much effort if I am the only one experiencing this.

This is bug, so we need to fix it. I have idea what is wrong and I reproduced it already:

Add new network printer via samba using user/pass auth. Finish yast module. Now 
/etc/cups/printers.conf is valid. Open module again (now string is internaly replaced to URL). Finish again - now /etc/cups/printers.conf is corrupted.

Excellent! Knowing what the problem is is 90% of solving it!

fixed, please check next build (it helps me if you check all about smb - create, edit and printing)

I presume this build will be part of the regular 10.1 updates? If so, then I will fully retest doing everything again.

Many thanks Michal.

Michal, please release for 10.1: MaintenanceTracker-5069

submited

Gents, this still is not fixed. It happened to me again today, and all I did was change "default" printer, etc.

I'm fully up to date on all rpm's as of today.

Which version of yast2-printer.rpm?

yast2-printer-2.13.13-6

By the way, I also find it wierd that KDE's printer config prints a great test print, but yast's seems to think the page is longer than it is. But that's another problem, for a later bug ;-)

Ugh,
the patch was created, but isn't available yet.
I'am sorry, but you must be patient and check it via online update.
But it is definitely fixed, so I'm closing this bug.

About test page - go ahead! Reporting a bug is first step to fix it ;-)

Michal,

Just to catch up regarding the KDE test print better than yast - I never followed up because I use a non-supported driver.  I have a Lexmark P3150 and I had to find and download a driver from the net and manually add it.  Given that its a crappy Lexmark, unsupported, using a 6 year old abandoned driver, I figured nobody should waste time on a bug report for it.