it is really good to see the http<>nntp gateway sweeping every ten
minutes, again…
gobs of thanks to whomever made that work, again!!
–
DD
Caveat-Hardware-Software
openSUSE®, the “German Engineered Automobiles” of operating systems!
On Fri, 22 Jul 2011 20:53:23 +0000, DenverD wrote:
> it is really good to see the http<>nntp gateway sweeping every ten
> minutes, again…
>
> gobs of thanks to whomever made that work, again!!
You’re welcome - Kim and I spent some time looking at whether this could
be reset now that we think we found the issue that was hanging it up and
addressed that, and switched it back early this week to see if it would
behave.
So far, so good. 
Jim
–
Jim Henderson
openSUSE Forums Administrator
Forum Use Terms & Conditions at http://tinyurl.com/openSUSE-T-C
On 2011-07-22 22:53, DenverD wrote:
> it is really good to see the http<>nntp gateway sweeping every ten minutes,
> again…
>
> gobs of thanks to whomever made that work, again!!
Good!
On which exact minutes does it kick? I need to adjust my cronjobs.
–
Cheers / Saludos,
Carlos E. R.
(from 11.4 x86_64 “Celadon” at Telcontar)
On Fri, 22 Jul 2011 23:50:06 +0000, Carlos E. R. wrote:
> On 2011-07-22 22:53, DenverD wrote:
>> it is really good to see the http<>nntp gateway sweeping every ten
>> minutes, again…
>>
>> gobs of thanks to whomever made that work, again!!
>
> Good!
>
> On which exact minutes does it kick? I need to adjust my cronjobs.
Any time ending with a ‘6’.
Jim
–
Jim Henderson
openSUSE Forums Administrator
Forum Use Terms & Conditions at http://tinyurl.com/openSUSE-T-C
On 2011-07-23 01:55, Jim Henderson wrote:
> On Fri, 22 Jul 2011 23:50:06 +0000, Carlos E. R. wrote:
>> On which exact minutes does it kick? I need to adjust my cronjobs.
>
> Any time ending with a ‘6’.
Aha. Then:
> -3,8,13,18,23,28,33,38,43,48,53,58 * * * * news /var/lib/news/bin/cronscriptparanovell
Thus I send on minutes ending on ‘3’ and receive on the '8’s.
–
Cheers / Saludos,
Carlos E. R.
(from 11.4 x86_64 “Celadon” at Telcontar)
On Sat, 23 Jul 2011 15:08:06 +0000, Carlos E. R. wrote:
> On 2011-07-23 01:55, Jim Henderson wrote:
>> On Fri, 22 Jul 2011 23:50:06 +0000, Carlos E. R. wrote:
>
>>> On which exact minutes does it kick? I need to adjust my cronjobs.
>>
>> Any time ending with a ‘6’.
>
> Aha. Then:
>
>
>> -3,8,13,18,23,28,33,38,43,48,53,58 * * * * news
>> /var/lib/news/bin/cronscriptparanovell
>
>
>
> Thus I send on minutes ending on ‘3’ and receive on the '8’s.
That should work OK, yes.
Jim
–
Jim Henderson
openSUSE Forums Administrator
Forum Use Terms & Conditions at http://tinyurl.com/openSUSE-T-C
On 2011-07-23, Carlos E. R. <robin_listas@no-mx.forums.opensuse.org> wrote:
> On 2011-07-23 01:55, Jim Henderson wrote:
>> On Fri, 22 Jul 2011 23:50:06 +0000, Carlos E. R. wrote:
>
>>> On which exact minutes does it kick? I need to adjust my cronjobs.
>>
>> Any time ending with a ‘6’. 
>
> Aha. Then:
>
>
>> -3,8,13,18,23,28,33,38,43,48,53,58 * * * * news /var/lib/news/bin/cronscriptparanovell
>
>
Thus I send on minutes ending on ‘3’ and receive on the '8’s.
I think you’re using leafnode, right?
There’s no need to wait so long to get your own posts back.
I run a fetch script that does a post-only first, waits 10 sec, then runs a
normal fetch. That 10 sec delay is enough to get your post back.
Running the post-only line if of course conditional to the presence of
outgoing posts.
#! /bin/bash
If the disk is active, fetch new txt news
if sudo /sbin/hdparm -C /dev/hdb | grep active | wc -l -ne 0 ]; then
If there is something to be posted, do that first
if ls "/var/spool/news/out.going/" | wc -l -ne 0 ]; then
Make a posting.
sudo -u news /usr/sbin/fetchnews -vP
echo
Wait long enough for the post be digested by the upstream server
echo sleeping 10 sec
sleep 10
echo
fi
Now fetch everything
sudo -u news /usr/sbin/fetchnews -vv
fi
You can remove the outside condition. I had just put that in so my server
would only fetch if his /dev/hdb was active. This partition holds /var/news
and is set to go to sleep mode after a few minutes of inactivity. In other
words, these frequent fetches only run if I use my newsreader. Otherwise it
keeps the partition dormant. There is of course also a daily (actually
nightly) unconditional fetch.
–
When in doubt, use brute force.
– Ken Thompson
On 2011-07-25 00:27, Rikishi42 wrote:
> On 2011-07-23, Carlos E. R. <> wrote:
>> Thus I send on minutes ending on ‘3’ and receive on the '8’s.
>
> I think you’re using leafnode, right?
Yep.
> There’s no need to wait so long to get your own posts back.
>
> I run a fetch script that does a post-only first, waits 10 sec, then runs a
> normal fetch. That 10 sec delay is enough to get your post back.
> Running the post-only line if of course conditional to the presence of
> outgoing posts.
The idea is not to get my own posts, but the posts of the rest of the
people
The thing is to send posts, as you do (thanks!) about two minutes before
the server gateway runs, then fetch about two minutes after. As the server
runs on the *6 minutes, I have to send on the *3 or *4, and fetch on the *8.
However, you do not need to do a send only run on the *3: a normal run does
send my posts and then fetch them on the same run. The upstream nntp server
is that fast.
> #! /bin/bash
>
> # If the disk is active, fetch new txt news
> if sudo /sbin/hdparm -C /dev/hdb | grep active | wc -l -ne 0 ]; then
neat
I wonder if there is a /proc entry with the same info :-?
> # If there is something to be posted, do that first
> if ls "/var/spool/news/out.going/" | wc -l -ne 0 ]; then
> # Make a posting.
> sudo -u news /usr/sbin/fetchnews -vP
> echo
> # Wait long enough for the post be digested by the upstream server
> echo sleeping 10 sec
> sleep 10
> echo
> fi
> # Now fetch everything
> sudo -u news /usr/sbin/fetchnews -vv
> fi
You can run a normal run instead, it will fetch your own mails with no wait.
> You can remove the outside condition. I had just put that in so my server
> would only fetch if his /dev/hdb was active. This partition holds /var/news
> and is set to go to sleep mode after a few minutes of inactivity. In other
> words, these frequent fetches only run if I use my newsreader. Otherwise it
> keeps the partition dormant. There is of course also a daily (actually
> nightly) unconditional fetch.
Very interesting!
I think there is another way: checking for a change in the
/var/spool/news/out.going/ timestamp. I don’t know how to do that without
awakening the disk, but tailf does something of the sort, it detects if a
file has grown without activating the disk, perhaps because it is cached.
They mention it on the manual:
tailf will print out the last 10 lines of a file and then
wait for the file to grow. It is similar to tail -f but
does not access the file when it is not growing. This has
the side effect of not updating the access time for the
file, so a filesystem flush does not occur periodically
when no log activity is happening.
–
Cheers / Saludos,
Carlos E. R.
(from 11.4 x86_64 “Celadon” at Telcontar)
On 2011-07-24, Carlos E. R. <robin_listas@no-mx.forums.opensuse.org> wrote:
> The idea is not to get my own posts, but the posts of the rest of the
> people
> The thing is to send posts, as you do (thanks!) about two minutes before
> the server gateway runs, then fetch about two minutes after. As the server
> runs on the *6 minutes, I have to send on the *3 or *4, and fetch on the *8.
Yep, I misread that a bit.
But then again, I don’t just fetch from the openSUSEe NNTP server, I also
have two other sources. So there was little point for me to have such
specific timeing.
When the system is active, the fetch runs every 15 min.
> However, you do not need to do a send only run on the *3: a normal run does
> send my posts and then fetch them on the same run. The upstream nntp server
> is that fast.
I’m pretty certain that hasn’t allways been the case. But my other sources
are not so fast, so I ended up with a compromise of a 10 sec wait.
>> #! /bin/bash
>>
>> # If the disk is active, fetch new txt news
>> if sudo /sbin/hdparm -C /dev/hdb | grep active | wc -l -ne 0 ]; then
>
> neat
>
> I wonder if there is a /proc entry with the same info :-?
Maybe, never checked. That server (soekris running CentOS) was setup a few
year ago and has been running without much change since.
It does the job, so…
–
When in doubt, use brute force.
– Ken Thompson
On 2011-07-28 22:00, Rikishi42 wrote:
> Maybe, never checked. That server (soekris running CentOS) was setup a few
> year ago and has been running without much change since.
> It does the job, so…
It is certainly a very interesting method
–
Cheers / Saludos,
Carlos E. R.
(from 11.4 x86_64 “Celadon” at Telcontar)