nntp<>http inoperative again--apparently

following the service outage and repair 19 to 21 June 2013 i have unsubscribed and resubscribed (twice) and find that (apparently) no post are flowing through the gateway in either direction…

i so report having just resubscribed and seeing (in my nntp client) a post in the hardware forum replied via nntp and that reply has not (yet) been seen in the http side…after waiting enough time for the gateway to have run twice, i then copied the message from the already delivered nntp post and pasted it into the http side…so, anyone can see both and have a look at fixing it…

the thread referenced is in: https://forums.opensuse.org/english/get-technical-help-here/hardware/488105-netgear-wna3100-wireless-usb-adapter.html#post2566693

i have subscribed to this thread and hope to soon see a test only email informing me to return the winky-blinky to read…


dd
ps: i intend to be SO patient as to be both invisible and unassuming any users/helpers deserve an estimated time to repair completion.

On Sun, 23 Jun 2013 09:16:02 +0000, DenverD wrote:

> following the service outage and repair 19 to 21 June 2013 i have
> unsubscribed and resubscribed (twice) and find that (apparently) no post
> are flowing through the gateway in either direction…

I’ve just checked and posts do seem to be flowing, but there’s something
weird about the numbering. New messages are coming in below the high
message counter, possibly filling in gaps from deleted messages. Doing a
re-fetch in pan of the last 14 days shows me the posts.

Try resetting your high message counters (I did this in pan by modifying
the newsrc-* files and removing all numbers at the end of the lines).

Jim

Jim Henderson
openSUSE Forums Administrator
Forum Use Terms & Conditions at http://tinyurl.com/openSUSE-T-C

This from the web side quoting my reply from the NNTP side.

Jim

On Sun, 23 Jun 2013 17:10:21 +0000, hendersj wrote:

> This from the web side quoting my reply from the NNTP side.

And this is from the NNTP side quoting my reply on the web side quoting
the NNTP side.

Jim

Jim Henderson
openSUSE Forums Administrator
Forum Use Terms & Conditions at http://tinyurl.com/openSUSE-T-C

On 06/23/2013 07:01 PM, Jim Henderson wrote:
> Try resetting your high message counters (I did this in pan by modifying
> the newsrc-* files and removing all numbers at the end of the lines).

i use thunderbird but found a
~/.thunderbird/[random].default/News/newsrc-support-forums.novell.com
which (despite being from that time prior to the http forums
existing) is being used still for the openSUSE nntp stuff…and,
removing the trailing numbers (and shutting down and restarting
TBird) i now see in the referenced thread both my http and nntp
posting…but, i do not see my nntp posting on the web side of that
thread…

but i do see my nntp post to my “unable to log out” at
http://forums.opensuse.org/showthread.php?t=488109 so, maybe it is
hit-and-miss, as Carlos mentions in his "after whatever happened…

i am SO glad that i don’t have to try to get it smooth again…i do
appreciate all the work that goes into supporting the newcomers
(including the nntp gateway that saves several of the helpers a LOT
of time)…


dd

On Sun, 23 Jun 2013 17:57:08 +0000, dd wrote:

> but, i do not see my nntp posting on the web side of that thread…

Ah, that might be due to the same reason why the readers are not picking
up new messages - the high message counter is set to ‘x’ and the new post
is ‘x-20’ or something.

I imagine those posts won’t come through the gateway, but it seems that
the reversion on the pointers is not too far back, so it should sort
itself out in a relatively short time, I imagine.

The forums I’ve looked at, the new posts are coming in within probably 50
of the latest high watermark.

Not an ideal situation, and I’ll talk about it with the tech guys to see
if there’s something we can do.

Jim

Jim Henderson
openSUSE Forums Administrator
Forum Use Terms & Conditions at http://tinyurl.com/openSUSE-T-C

On 2013-06-23 19:01, Jim Henderson wrote:
> On Sun, 23 Jun 2013 09:16:02 +0000, DenverD wrote:
>
>> following the service outage and repair 19 to 21 June 2013 i have
>> unsubscribed and resubscribed (twice) and find that (apparently) no post
>> are flowing through the gateway in either direction…
>
> I’ve just checked and posts do seem to be flowing, but there’s something
> weird about the numbering. New messages are coming in below the high
> message counter, possibly filling in gaps from deleted messages. Doing a
> re-fetch in pan of the last 14 days shows me the posts.
>
> Try resetting your high message counters (I did this in pan by modifying
> the newsrc-* files and removing all numbers at the end of the lines).
>
> Jim

I was asking about this on another thread:

Newsgroups: opensuse.org.feedback.forums.support-information
Subject: Re: NNTP access (again)?

+++·······································
I tried again with the thread “openSuse 12.3 kde different cursor theme
at login screen” on the application forum - using pan directly connected
to nntp.opensuse.org. I opened the group, and I could not see it. The
groups are set to download new posts on entry. I resubscribed several
times already, no difference.

What it works is telling Pan to download ALL headers, not new headers
only. Or to download the headers of the last 120 days. THEN I get that
thread I mentioned, started today, and others.

On Thunderbird I do not see it, working via leafnode, so leafnode hasn’t
got that post either. In Thunderbird I resubsribed already to all the
groups, no difference.

What works in leafnode, is go to the
/var/spool/news/leaf.node/nntp.opensuse.org file, edit it, and edit the
line from

opensuse.org.help.applications 161733

to

opensuse.org.help.applications 161000

then run fetchnews again. This is basically the same as telling Pan to
download the last 100 days headers.

In the log I then see:


> <7.7> 2013-06-23 20:00:02 Telcontar fetchnews 12662 - -  >XOVER 161000-161732
> <7.7> 2013-06-23 20:00:03 Telcontar fetchnews 12662 - -  <224 Overview Information Follows
> <7.7> 2013-06-23 20:00:03 Telcontar fetchnews 12662 - -  opensuse.org.help.applications: killed 161000 (<chief_sealth.5wceao@no-mx.forums.opensuse.org>), already fetched before
> <7.7> 2013-06-23 20:00:03 Telcontar fetchnews 12662 - -  opensuse.org.help.applications: killed 161001 (<caf4926.5wceap@no-mx.forums.opensuse.org>), already fetched before
> ...
> <7.7> 2013-06-23 20:00:08 Telcontar fetchnews 12662 - -  opensuse.org.help.applications: killed 161583 (<NikTh.5xchdc@no-mx.forums.opensuse.org>), already fetched before
(**)
> <7.7> 2013-06-23 20:00:08 Telcontar fetchnews 12662 - -  opensuse.org.help.applications: killed 161626 (<wolfi323.5x2nnz@no-mx.forums.opensuse.org>), already fetched before
> ...
> <7.7> 2013-06-23 20:00:09 Telcontar fetchnews 12662 - -  opensuse.org.help.applications: killed 161731 (<tsu2.5x6to1@no-mx.forums.opensuse.org>), already fetched before
> <7.7> 2013-06-23 20:00:09 Telcontar fetchnews 12662 - -  opensuse.org.help.applications: killed 161732 (<vazhavandan.5x72go@no-mx.forums.opensuse.org>), already fetched before
> <7.7> 2013-06-23 20:00:09 Telcontar fetchnews 12662 - -  >ARTICLE 161584
> <7.7> 2013-06-23 20:00:09 Telcontar fetchnews 12662 - -  >ARTICLE 161585
> ...
> <7.7> 2013-06-23 20:00:26 Telcontar fetchnews 12662 - -  >ARTICLE 161624
> <7.7> 2013-06-23 20:00:26 Telcontar fetchnews 12662 - -  >ARTICLE 161625
> <7.6> 2013-06-23 20:00:27 Telcontar fetchnews 12662 - -  opensuse.org.help.applications: 42 articles fetched (to 72830), 680 killed
> <7.7> 2013-06-23 20:00:27 Telcontar fetchnews 12662 - -  >GROUP opensuse.org.help.programming-scripting.build-service

The highwater mark reads now “161733”, but notice that the last post I
downloaded was only “161625”, so no new posts will be retrieved
automatically, that mark is wrong.

The server does indeed have post 161733 (see log above), but it
corresponds to a post that has already been retrieved before. The posts
I miss were somewhere before that, from 161584 to 161625. Notice the
missing section marked with “(**)” above? Those are note lines I
removed, that’s a real jump at the server, those are the posts that were
not downloaded before, that were skipped because they are below the high
water mark.

Could the server be filling the holes of the posts that were not sent
during these days? Not the posts, but the numbers, like slots. No, this
makes no sense…
·······································+±

Or does it?


Cheers / Saludos,

Carlos E. R.
(from 12.3 x86_64 “Dartmouth” at Telcontar)

On 2013-06-24 03:33, Carlos E. R. wrote:

> What works in leafnode, is go to the
> /var/spool/news/leaf.node/nntp.opensuse.org file, edit it, and edit the
> line from
>
> opensuse.org.help.applications 161733
>
> to
>
> opensuse.org.help.applications 161000
>
>
> then run fetchnews again. This is basically the same as telling Pan to
> download the last 100 days headers.

Easier:


/usr/sbin/fetchnews -v -x 5000

From the manual:


-x number
Go  number  articles  back and try to re-fetch
what's missing. This can be useful to
pull data from broken servers that have
either overview and group data out of  synch
or create articles out of article number order.

The problem is not knowing how many articles to go back. 5000 is a
brute, but it works.


Cheers / Saludos,

Carlos E. R.
(from 12.3 x86_64 “Dartmouth” at Telcontar)

FWIW, if you’re using knode(kontact), you can go to ~/.kde4/share/apps/knode, and edit each of the *.grpinfo files. Subtract something from the value on the ‘lastMsg’ line, and the articles will appear.

I’m getting a lot of “connection refused” notices from both servers. A
sample:


> <7.4> 2013-06-24 19:13:05 Telcontar fetchnews 15526 - -  warning: nntp.opensuse.org: connection to 130.57.5.118 failed: Connection refused
> <7.4> 2013-06-24 19:13:05 Telcontar fetchnews 15526 - -  warning: nntp.novell.com: connection to 130.57.4.24 failed: Connection refused
> <7.4> 2013-06-24 19:13:05 Telcontar fetchnews 15526 - -  WARNING: some servers have not been queried!
> <7.4> 2013-06-24 19:18:05 Telcontar fetchnews 15677 - -  warning: nntp.opensuse.org: connection to 130.57.5.118 failed: Connection refused
> <7.4> 2013-06-24 19:18:06 Telcontar fetchnews 15677 - -  warning: nntp.novell.com: connection to 130.57.4.24 failed: Connection refused
> <7.4> 2013-06-24 19:18:06 Telcontar fetchnews 15677 - -  WARNING: some servers have not been queried!
> <7.4> 2013-06-24 19:23:05 Telcontar fetchnews 15776 - -  warning: nntp.opensuse.org: connection to 130.57.5.118 failed: Connection refused
> <7.4> 2013-06-24 19:23:05 Telcontar fetchnews 15776 - -  warning: nntp.novell.com: connection to 130.57.4.24 failed: Connection refused
> <7.4> 2013-06-24 19:23:05 Telcontar fetchnews 15776 - -  WARNING: some servers have not been queried!


Cheers / Saludos,

Carlos E. R.
(from 12.3 x86_64 “Dartmouth” at Telcontar)

On Mon, 24 Jun 2013 01:33:08 +0000, Carlos E. R. wrote:

> Could the server be filling the holes of the posts that were not sent
> during these days? Not the posts, but the numbers, like slots. No, this
> makes no sense… ·······································+±
>
> Or does it?

Yes, that’s more or less exactly what’s happening.

As I explained in another thread (and maybe we should get this all to a
single thread), there are two servers that are synchronized by a third.
The third is the one that failed. Meanwhile, posts from the NNTP side
going to one or the other of the NNTP servers caused the message counters
between the two servers to get out of sync.

That issue is being worked on so it won’t continue to be a problem.
Going back (as I suggested elsewhere) and fixing from the point of the
failure is a non-trivial problem, so probably won’t be fixed, but the
servers’ high message counters are (I believe) going to be brought
current so the problem doesn’t continue.

Jim


Jim Henderson
openSUSE Forums Administrator
Forum Use Terms & Conditions at http://tinyurl.com/openSUSE-T-C

On 2013-06-24 19:57, Jim Henderson wrote:
> On Mon, 24 Jun 2013 01:33:08 +0000, Carlos E. R. wrote:
>
>> Could the server be filling the holes of the posts that were not sent
>> during these days? Not the posts, but the numbers, like slots. No, this
>> makes no sense… ·······································+±
>>
>> Or does it?
>
> Yes, that’s more or less exactly what’s happening.
>
> As I explained in another thread (and maybe we should get this all to a
> single thread),

I must have missed that one. Please, could you say tittle and group, so
that I can pick it up?

> there are two servers that are synchronized by a third.
> The third is the one that failed.

Oh.

> Meanwhile, posts from the NNTP side
> going to one or the other of the NNTP servers caused the message counters
> between the two servers to get out of sync.

I think I see.

> That issue is being worked on so it won’t continue to be a problem.
> Going back (as I suggested elsewhere) and fixing from the point of the
> failure is a non-trivial problem, so probably won’t be fixed, but the
> servers’ high message counters are (I believe) going to be brought
> current so the problem doesn’t continue.

Does this affect our posts getting on the web side?


Cheers / Saludos,

Carlos E. R.
(from 12.3 x86_64 “Dartmouth” at Telcontar)

On 2013-06-24 22:33, Carlos E. R. wrote:
> On 2013-06-24 19:57, Jim Henderson wrote:

>> As I explained in another thread (and maybe we should get this all to a
>> single thread),
>
> I must have missed that one. Please, could you say tittle and group, so
> that I can pick it up?

Never mind, found it. The one on the announcement forum; I’m tracking
it, but I saw this one first.


Cheers / Saludos,

Carlos E. R.
(from 12.3 x86_64 “Dartmouth” at Telcontar)

On Mon, 24 Jun 2013 20:33:06 +0000, Carlos E. R. wrote:

>> That issue is being worked on so it won’t continue to be a problem.
>> Going back (as I suggested elsewhere) and fixing from the point of the
>> failure is a non-trivial problem, so probably won’t be fixed, but the
>> servers’ high message counters are (I believe) going to be brought
>> current so the problem doesn’t continue.
>
> Does this affect our posts getting on the web side?

It shouldn’t now.

Jim


Jim Henderson
openSUSE Forums Administrator
Forum Use Terms & Conditions at http://tinyurl.com/openSUSE-T-C

DenverD wrote:
>
> following the service outage and repair 19 to 21 June 2013 i have
> unsubscribed and resubscribed (twice) and find that (apparently) no post
> are flowing through the gateway in either direction…
>
> i so report having just resubscribed and seeing (in my nntp client) a
> post in the hardware forum replied via nntp and that reply has not (yet)
> been seen in the http side…after waiting enough time for the gateway
> to have run twice, i then copied the message from the already delivered
> nntp post and pasted it into the http side…so, anyone can see both
> and have a look at fixing it…
>
> the thread referenced is in: http://tinyurl.com/jweon3d
>
> i have subscribed to this thread and hope to soon see a test only email
> informing me to return the winky-blinky to read…
>

Testing nntp

> Testing nntp

loud and clear.


dd

I have a few queries regarding nntp
==> Will forums accept any picture attachments i send from SeaMonkey Newsgroups account ?
==> Is there a sign in mechanism for nntp . There seems to be no validation mechanism :frowning:
==> Should i reply to “newsgroup” or “individual” . Also there is reply to all button . How will reply to all work ?
==> will bbcode work ?
Any links to a detailed guide would also be cool

On 07/18/2013 07:06 AM, vazhavandan wrote:
>
> I have a few queries regarding nntp

this is not a help forum and this thread has nothing to do with your
questions…

so, to make the answers available to all those who follow, i’d
suggest you post your questions to a new thread with a subject maybe
like “How to use NNTP with the openSUSE Forums?” and place that in
the “How to use the forums?” or the forums.support-information
newsgroup…


dd

On 7/18/2013 12:06 AM, vazhavandan wrote:
>
> I have a few queries regarding nntp
> ==> Will forums accept any picture attachments i send from SeaMonkey
> Newsgroups account ?
> ==> Is there a sign in mechanism for nntp . There seems to be no
> validation mechanism :frowning:
> ==> Should i reply to “newsgroup” or “individual” . Also there is reply
> to all button . How will reply to all work ?
> ==> will bbcode work ?
> Any links to a detailed guide would also be cool
>
>
vazhavandan;

  1. AFAIK you cannot send pictures, but you can send a URL for the picture.

  2. There is no need to authenticate from nntp. This has been discussed before.

  3. Reply to “newsgroup” there is generally no individual address to send to; just something like:
    vazhavandan@no-mx.forums.opensuse.org

  4. For BBcodes see:
    http://forums.opensuse.org/misc.php?do=bbcode
    and you should read:
    http://forums.opensuse.org/faq.php?faq=novfor#faq_nntp


P.V.
“We’re all in this together, I’m pulling for you” Red Green

On 2013-07-18 07:06, vazhavandan wrote:
>
> I have a few queries regarding nntp

Better on a new thread. Let me start one for you, because this would be
a thread hijack.

newsgroup: opensuse.org.feedback.forums.support-information
subject: How to use NNTP with the openSUSE Forums?


Cheers / Saludos,

Carlos E. R.
(from 12.3 x86_64 “Dartmouth” at Telcontar)