SEO

Here is the situation:

I have a friend with the following system;
E35
HP 10.2
LVM

Several days ago he filled up the filesystem. After deleting some files he
found that the lp system didn't work. In particular, he would run lpsched
and get the "Scheduler is running' message. However, when he did an lpstat -t
he got a "Schedular is off" message. Also if he did a ps -ef the scheduler
(lpsched) is not running. Nothing is being written to the lpsched log files.

Clearly, printing is impossible.

We have gone as far as to do a restore from backup of the system from several
days previous to the problem. We have also done hard reboots of the system.

I am completely stymied on how to find and fix the problem. Any suggestions?





--
--
Larry Headlund lmh@world.std.com Mathematical Engineering, Inc.
(617) 242 7741
Unix, X and Motif Consulting Speaking for myself at most.

lmh@TheWorld.com (Larry M Headlund) wrote:



>Here is the situation:
>
>I have a friend with the following system;
> E35
> HP 10.2
> LVM
>
>Several days ago he filled up the filesystem. After deleting some files he
>found that the lp system didn't work. In particular, he would run lpsched
>and get the "Scheduler is running' message. However, when he did an lpstat -t
>he got a "Schedular is off" message. Also if he did a ps -ef the scheduler
>(lpsched) is not running. Nothing is being written to the lpsched log files.
>
>Clearly, printing is impossible.
>
>We have gone as far as to do a restore from backup of the system from several
>
>days previous to the problem. We have also done hard reboots of the system.
>
>I am completely stymied on how to find and fix the problem. Any suggestions?
>
>
>
>
>
>--
>--
>Larry Headlund lmh@world.std.com Mathematical Engineering, Inc.
> (617) 242 7741
>Unix, X and Motif Consulting Speaking for myself at most.

Larry,

Do /var/spool/lp/SCHEDLOCK and FIFO still exist? They should not be there when
the scheduler is down.

Eric

In article <20050115134933.08487.00000022@mb-m06.aol.com>,
ECStahl <ecstahl@aol.com> wrote:
>lmh@TheWorld.com (Larry M Headlund) wrote:
>>Here is the situation:
>>
>>I have a friend with the following system;
>> E35
>> HP 10.2
>> LVM
>>
>>Several days ago he filled up the filesystem. After deleting some files he
>>found that the lp system didn't work. In particular, he would run lpsched
>>and get the "Scheduler is running' message. However, when he did an lpstat -t
>>he got a "Schedular is off" message. Also if he did a ps -ef the scheduler
>>(lpsched) is not running. Nothing is being written to the lpsched log files.
>>
>>Clearly, printing is impossible.
>>
>>We have gone as far as to do a restore from backup of the system from several
>>
>>days previous to the problem. We have also done hard reboots of the system.
>>
>>I am completely stymied on how to find and fix the problem. Any suggestions?
>Larry,
>
>Do /var/spool/lp/SCHEDLOCK and FIFO still exist? They should not be there when
>the scheduler is down.

Indeed the two files are there and also CLD_FIFO. As a test I just now
removed those three files, tried to restart lpsched and still no joy.
They get recreated each time I run lpsched but the program dies sometime
thereafter.


--
--
Larry Headlund lmh@world.std.com Mathematical Engineering, Inc.
(617) 242 7741
Unix, X and Motif Consulting Speaking for myself at most.

Larry M Headlund wrote:

> In article <20050115134933.08487.00000022@mb-m06.aol.com>,
> ECStahl <ecstahl@aol.com> wrote:
>
>>lmh@TheWorld.com (Larry M Headlund) wrote:
>>
>>>Here is the situation:
>>>
>>>I have a friend with the following system;
>>> E35
>>> HP 10.2
>>> LVM
>>>
>>>Several days ago he filled up the filesystem. After deleting some files he
>>>found that the lp system didn't work. In particular, he would run lpsched
>>>and get the "Scheduler is running' message. However, when he did an lpstat -t
>>>he got a "Schedular is off" message. Also if he did a ps -ef the scheduler
>>>(lpsched) is not running. Nothing is being written to the lpsched log files.
>>>
>>>Clearly, printing is impossible.
>>>
>>>We have gone as far as to do a restore from backup of the system from several
>>>
>>>days previous to the problem. We have also done hard reboots of the system.
>>>
>>>I am completely stymied on how to find and fix the problem. Any suggestions?
>>
>>Larry,
>>
>>Do /var/spool/lp/SCHEDLOCK and FIFO still exist? They should not be there when
>>the scheduler is down.
>
>
> Indeed the two files are there and also CLD_FIFO. As a test I just now
> removed those three files, tried to restart lpsched and still no joy.
> They get recreated each time I run lpsched but the program dies sometime
> thereafter.
>
>
stupid question, is LP=1 in /etc/rc.config.d/lp?

In article <zGjGd.49644$kq2.39011@twister.nyc.rr.com>,
Alan D Johnson <adjtech@hvc.rr.com> wrote:
>>>
>>>Do /var/spool/lp/SCHEDLOCK and FIFO still exist? They should not be there when
>>>the scheduler is down.
>>
>>
>> Indeed the two files are there and also CLD_FIFO. As a test I just now
>> removed those three files, tried to restart lpsched and still no joy.
>> They get recreated each time I run lpsched but the program dies sometime
>> thereafter.
>>
>>
>stupid question, is LP=1 in /etc/rc.config.d/lp?

Yes LP=1 in the config files.

I just had my friend do a reboot and we get the following behavior:
If remove the three named pipe files they are NOT recreated when
lpsched attempts to restart.
With the three named pipes removed lpstat -t gives the message
"Can't open printer status file."


--
--
Larry Headlund lmh@world.std.com Mathematical Engineering, Inc.
(617) 242 7741
Unix, X and Motif Consulting Speaking for myself at most.

Larry M Headlund wrote:
> In article <zGjGd.49644$kq2.39011@twister.nyc.rr.com>,
> Alan D Johnson <adjtech@hvc.rr.com> wrote:
>
>>>>Do /var/spool/lp/SCHEDLOCK and FIFO still exist? They should not be there when
>>>>the scheduler is down.
>>>
>>>
>>>Indeed the two files are there and also CLD_FIFO. As a test I just now
>>>removed those three files, tried to restart lpsched and still no joy.
>>>They get recreated each time I run lpsched but the program dies sometime
>>>thereafter.
>>>
>>>
>>
>>stupid question, is LP=1 in /etc/rc.config.d/lp?
>
>
> Yes LP=1 in the config files.
>
> I just had my friend do a reboot and we get the following behavior:
> If remove the three named pipe files they are NOT recreated when
> lpsched attempts to restart.
> With the three named pipes removed lpstat -t gives the message
> "Can't open printer status file."
>
>
has anyone changed any perms? /var/spool/lp should be owned by lp:bin.
Is it trying to start at all, and does it core? If you did a reboot what
is the message in /etc/rc.log when the system started the lp daemon?

In article <idmGd.58986$ld2.21538169@twister.nyc.rr.com>,
Alan D Johnson <adjtech@hvc.rr.com> wrote:
>Larry M Headlund wrote:
>> In article <zGjGd.49644$kq2.39011@twister.nyc.rr.com>,
>> Alan D Johnson <adjtech@hvc.rr.com> wrote:
>>
>>>>>Do /var/spool/lp/SCHEDLOCK and FIFO still exist? They should not be there when
>>>>>the scheduler is down.
>>>>
>>>>
>>>>Indeed the two files are there and also CLD_FIFO. As a test I just now
>>>>removed those three files, tried to restart lpsched and still no joy.
>>>>They get recreated each time I run lpsched but the program dies sometime
>>>>thereafter.
>>>>
>>>>
>>>
>>>stupid question, is LP=1 in /etc/rc.config.d/lp?
>>
>>
>> Yes LP=1 in the config files.
>>
>> I just had my friend do a reboot and we get the following behavior:
>> If remove the three named pipe files they are NOT recreated when
>> lpsched attempts to restart.
>> With the three named pipes removed lpstat -t gives the message
>> "Can't open printer status file."
>>
>>
>has anyone changed any perms? /var/spool/lp should be owned by lp:bin.
>Is it trying to start at all, and does it core? If you did a reboot what
>is the message in /etc/rc.log when the system started the lp daemon?

Owner ships are as they should be. It does not appear to core, at least
there are no files in te system named core as you wopuld expect from
a failure like that.

The relevant /etc/rc.log appears normal.

Now, my friend just told me (why did he wait) that /usr was reporting
zero free! (LVM) even though there was free space and even after we
deleted some additional files.. I did the usual juggling to clean out
the filesystem, did a mkfs on the relevant volume, copied back in the files
for /usr but still no joy.

My next step is going to be to clean out the /usr volume again and
restore just that directory from a backup.

Bye the way, thank you for your continuing help!

--
--
Larry Headlund lmh@world.std.com Mathematical Engineering, Inc.
(617) 242 7741
Unix, X and Motif Consulting Speaking for myself at most.

Larry M Headlund wrote:
> In article <idmGd.58986$ld2.21538169@twister.nyc.rr.com>,
> Alan D Johnson <adjtech@hvc.rr.com> wrote:
>
>>Larry M Headlund wrote:
>>
>>>In article <zGjGd.49644$kq2.39011@twister.nyc.rr.com>,
>>>Alan D Johnson <adjtech@hvc.rr.com> wrote:
>>>
>>>
>>>>>>Do /var/spool/lp/SCHEDLOCK and FIFO still exist? They should not be there when
>>>>>>the scheduler is down.
>>>>>
>>>>>
>>>>>Indeed the two files are there and also CLD_FIFO. As a test I just now
>>>>>removed those three files, tried to restart lpsched and still no joy.
>>>>>They get recreated each time I run lpsched but the program dies sometime
>>>>>thereafter.
>>>>>
>>>>>
>>>>
>>>>stupid question, is LP=1 in /etc/rc.config.d/lp?
>>>
>>>
>>>Yes LP=1 in the config files.
>>>
>>>I just had my friend do a reboot and we get the following behavior:
>>> If remove the three named pipe files they are NOT recreated when
>>>lpsched attempts to restart.
>>> With the three named pipes removed lpstat -t gives the message
>>>"Can't open printer status file."
>>>
>>>
>>
>>has anyone changed any perms? /var/spool/lp should be owned by lp:bin.
>>Is it trying to start at all, and does it core? If you did a reboot what
>>is the message in /etc/rc.log when the system started the lp daemon?
>
>
> Owner ships are as they should be. It does not appear to core, at least
> there are no files in te system named core as you wopuld expect from
> a failure like that.
>
> The relevant /etc/rc.log appears normal.
>
> Now, my friend just told me (why did he wait) that /usr was reporting
> zero free! (LVM) even though there was free space and even after we
> deleted some additional files.. I did the usual juggling to clean out
> the filesystem, did a mkfs on the relevant volume, copied back in the files
> for /usr but still no joy.
>
> My next step is going to be to clean out the /usr volume again and
> restore just that directory from a backup.
>
> Bye the way, thank you for your continuing help!
>
is /usr/spool still a link to /var/spool? if you whacked the fs and then
restored it it may have put it back as a directory, not a link.

In article <SUvGd.44766$Yh2.20514122@twister.nyc.rr.com>,
Alan D Johnson <adjtech@hvc.rr.com> wrote:
>>
>is /usr/spool still a link to /var/spool? if you whacked the fs and then
>restored it it may have put it back as a directory, not a link.

Yes, /usr/spool is a link to /var/spool as it should be.


--
--
Larry Headlund lmh@world.std.com Mathematical Engineering, Inc.
(617) 242 7741
Unix, X and Motif Consulting Speaking for myself at most.

Hi Larry

I sent to your mail address instructions for the so called spoolkick
procedure, and some other hints for clearing the printer system.

Florian



--
mail an "fanwander AT mnet MINUS online PUNKT de"