Gateway slow

ruby-talk
1

Hi there,

apparently the direction news -> mailing is broken (or extremely
slow). I don't see a posting I did ~5 hours ago. Anybody with
similar experiences?

Cheers

robert

···

--
remember.guy do |as, often| as.you_can - without end
http://blog.rubybestpractices.com/

2

Can you point me at the post you are referring to on the newsgroup side so I can look it up and see what happened to it? Thanks.

James Edward Gray II

···

On Mar 19, 2010, at 7:20 AM, Robert Klemme wrote:

apparently the direction news -> mailing is broken (or extremely
slow). I don't see a posting I did ~5 hours ago. Anybody with
similar experiences?

3

Here's a test posted to Google groups at 07:26 MST.

···

On Mar 19, 6:20 am, Robert Klemme <shortcut...@googlemail.com> wrote:

apparently the direction news -> mailing is broken (or extremely
slow). I don't see a posting I did ~5 hours ago. Anybody with
similar experiences?

4

http://groups.google.de/group/comp.lang.ruby/browse_frm/thread/1030ad4858886ddf/af74b26045f69573

My first posting - I don't see it on the mailing list.

Thanks for offering support!

Kind regards

robert

···

2010/3/19 James Edward Gray II <james@graysoftinc.com>:

On Mar 19, 2010, at 7:20 AM, Robert Klemme wrote:

apparently the direction news -> mailing is broken (or extremely
slow). I don't see a posting I did ~5 hours ago. Anybody with
similar experiences?

Can you point me at the post you are referring to on the newsgroup side so I can look it up and see what happened to it? Thanks.

--
remember.guy do |as, often| as.you_can - without end
http://blog.rubybestpractices.com/

5

Fred, we hare having trouble getting Usenet postings from your server. The issue comes when I try to skip forward to a recent article, using this code in the gateway:

# switch to the last message we sent (or the first real message before that)
begin
  Timeout.timeout(30) do
    nntp.stat(last_id_sent)
  end
rescue Timeout::Error
    log.error "The NNTP message shift timed out."
    exit
rescue
  if $!.message.include?("Bad article number") and last_id_sent.nonzero?
    last_id_sent -= 1
    retry
  else
    log.fatal "Unable to switch to the last message: #{$!.message}"
    exit -1
  end
end

The server appears to be denying all of my stat() calls as last_id_sent just keeps counting down:

# ruby /var/ruby_talk_gateway/bin/news_to_mail.rb /var/log/gateway_news_to_mail.log
263164
263163
263162
263161
263160
263159
263158
263157
263156
…

Any thoughts?

James Edward Gray II

···

On Mar 19, 2010, at 10:29 AM, Robert Klemme wrote:

2010/3/19 James Edward Gray II <james@graysoftinc.com>:

On Mar 19, 2010, at 7:20 AM, Robert Klemme wrote:

apparently the direction news -> mailing is broken (or extremely
slow). I don't see a posting I did ~5 hours ago. Anybody with
similar experiences?

Can you point me at the post you are referring to on the newsgroup side so I can look it up and see what happened to it? Thanks.

http://groups.google.de/group/comp.lang.ruby/browse_frm/thread/1030ad4858886ddf/af74b26045f69573

My first posting - I don't see it on the mailing list.

6

Hi.

Can you point me at the post you are referring to on the newsgroup side so I
can look it up and see what happened to it? Thanks.

http://groups.google.de/group/comp.lang.ruby/browse_frm/thread/1030ad4858886ddf/af74b26045f69573

My first posting - I don't see it on the mailing list.

Fred, we hare having trouble getting Usenet postings from your server.
The issue comes when I try to skip forward to a recent article, using
this code in the gateway:

/.../

The server appears to be denying all of my stat() calls as last_id_sent
just keeps counting down:

# ruby /var/ruby_talk_gateway/bin/news_to_mail.rb /var/log/gateway_news_to_mail.log
263164
263163
263162
263161
263160
263159
263158
263157
263156
?

Any thoughts?

I'm not sure of what this code is supposed to do. Well, of why it does
that, to be precise - are you trying to find the last gated article to
retrieve an ID ?

In this case, the reason it's looping is quite simply that you are
trying to get articles below the first article in the group (returned by
the "group" command).

11:00 news@talisker:~% telnet localhost 119
/.../
group comp.lang.ruby
211 204 263180 263385 comp.lang.ruby

          ^^^^^^ ^^^^^^
          lomark himark
          

stat 263180
223 263180 <dd1a5173-0e1f-49f8-9081-0e4db0f0ee90@l25g2000yqd.googlegroups.com> status
stat 263179
423 Bad article number

This is due to yet another hardware problem with the box recently (the
disk controller died). It seems the overview database (the database of
important header fields) has been corrupted for this group, leading to a
very low number of articles (205) while I have a few thousands on disk.

I have just launched a rebuild of the databases, it should recover all
the articles... but it's a bit slow on that machine.

The good news is that I have found a replacement server. I'll try to
put it in production in a few weeks.

CYA,

Fred

···

Le 19 mars 2010 à 17:00, James Edward Gray II a écrit :
--
It enters your veins It enters your soul It tries to obsess you
It looks for defaults Try not to lose Not to lose control
It tries to destroy you It tries to tear you down
                                            (Hooverphonic, Plus Profond)

7

Much better !

Fred

···

Le 20 mars 2010 à 11:28, F. Senault a écrit :

11:00 news@talisker:~% telnet localhost 119
/.../
group comp.lang.ruby
211 204 263180 263385 comp.lang.ruby

          ^^^^^^ ^^^^^^
          lomark himark
group comp.lang.ruby
211 4437 259122 263387 comp.lang.ruby

--
The quality of life and the world in general become much better if we
start chopping off spammers' fingers at a rate of say, 1 finger per
message, with bonus pre-amputation limb flattening for html or other
"prettification" mail. (Peter N. M. Hansteen in the SDM)

8

Testing a post from the Usenet side.

James Edward Gray II

···

On Mar 20, 5:57 am, "F. Senault" <f...@lacave.net> wrote:

Le 20 mars 2010 à 11:28, F. Senault a écrit :

>> 11:00 news@talisker:~% telnet localhost 119
>> /.../
>> group comp.lang.ruby
>> 211 204 263180 263385 comp.lang.ruby
> ^^^^^^ ^^^^^^
> lomark himark

> group comp.lang.ruby
> 211 4437 259122 263387 comp.lang.ruby

Much better !

9

Yep, it looks like we are back in business. Thanks!

James Edward Gray II

···

On Mar 20, 2010, at 7:42 AM, F. Senault wrote:

Le 20 mars 2010 à 11:28, F. Senault a écrit :

> 11:00 news@talisker:~% telnet localhost 119
> /.../
> group comp.lang.ruby
> 211 204 263180 263385 comp.lang.ruby
         ^^^^^^ ^^^^^^
         lomark himark

> group comp.lang.ruby
> 211 4437 259122 263387 comp.lang.ruby

Much better !

10

Thanks from me as well!

Kind regards

  robert

···

On 03/20/2010 09:52 PM, James Edward Gray II wrote:

On Mar 20, 2010, at 7:42 AM, F. Senault wrote:

Le 20 mars 2010 à 11:28, F. Senault a écrit :

> 11:00 news@talisker:~% telnet localhost 119
> /.../
> group comp.lang.ruby
> 211 204 263180 263385 comp.lang.ruby
         ^^^^^^ ^^^^^^
         lomark himark

> group comp.lang.ruby
> 211 4437 259122 263387 comp.lang.ruby

Much better !

Yep, it looks like we are back in business. Thanks!

--
remember.guy do |as, often| as.you_can - without end
http://blog.rubybestpractices.com/