Re: Server connection lost -> "set prefer_head 1" solves it

* John E. Davis <[email protected]> [2008-08-29 10:27]:
> Sven Guckes <[email protected]> wrote:
> > slrn seems to hang at these lines:
> >
> >   514 write(3, "XHDR Path\r\n", 11)           = 11
> >   600 write(3, "XHDR Path\r\n", 11)           = 11
> >   686 write(3, "XHDR Path\r\n", 11)           = 11
>
> Please try pre0.9.9p1-12, where the probe for XHDR is made earlier.

a new install?  hmm... no sure how i can get this version fast.

> If you still see timeouts then my
> only suggestion is that you put
>    set prefer_head 1
> in your .slrnrc file.

tried this - worked! :-)
i now have a fast startup again
just like i remember slrn. yay!

now i need to understand *why* this is.
how much difference is there between
this XHDR and "fetching headers"?

just for the record - here is what is
available in changes.txt and manual.txt:

  changes.txt:
  35. Expensive scoring should now be faster on servers that support XHDR. If
      you find the opposite to be true for you, set "prefer_head" to 1.

  manual.txt:
  6.74.  prefer_head - Type: integer - Default: 0
  This variable affects "expensive scoring": If available, slrn uses the
  XOVER and XHDR extensions to retrieve only the needed header lines;
  otherwise, the HEAD NNTP command is used to get all headers.
  Although this notably increases network traffic, it might be faster,
  depending on your connection and the number of "expensive" headers
  you score on.  In this case, you can set this variable and
  slrn will prefer HEAD over XOVER / XHDR.
  When set to 2, slrn never tries to use XOVER.
  This is only useful for servers
  with broken XOVER implementations.

> > there are a LOT of stats on /etc/localtime -
> > the complete output of 2460 lines contains
> > 464 stats on /etc/localtime alone,
> > mostly a dozen times in a block.
>
> Although I was unable to reproduce this, I suspect that
> the culprit is the strftime function, which may "stat"
> /etc/localtime to see if the time-zone has changed.
> This can happen if you have a laptop and are on the road.

i log in from a laptop to a server -
but i use slrn from that server.
i find it strange that slrn would check
/etc/localtime a dozen times in a row.
why would it do this so often?  hmm..

> This behavior appears be libc dependent.
> for more information see:
> <http://sourceware.org/bugzilla/show_bug.cgi?id=5184>

  Additional Comment #1 From Ulrich Drepper  2007-10-16 05:44

  No, /etc/localtime is not touched too often.  All the time functions are
  required to check TZ all the time.  Missing that envvar, /etc/localtime is used.
  People who move laptops from one timezone to the other appreciate that.
  If this is no problem of yours just set TZ to /etc/localtime and all this will stop.

so... setting it how exactly?  example?

Sven

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/