[Bioperl-l] Speed of bioperl mailing list distribution

Jason Stajich jason.stajich at duke.edu
Sat Apr 22 09:11:53 EDT 2006


(ChrisD will probably be offline for a few days with some server  
issues for his company.)

One reason for the slowdown can be greylisting.  See http:// 
en.wikipedia.org/wiki/Greylisting  - We have enabled greylisting so  
mails to the list from new IP addresses will take longer the first  
time a message is sent because the server requires the sender's SMTP  
server to reconnect to the OBF server a second time.  This cuts down  
dramatically on spam that is sent by transient IPs or by spammers and  
we feel that the short delay is better than the reams of spam we had  
to moderate in the past.  The CVS commits come from an internal  
server so they will be already okayed without a retry.  Posts from  
non-subscribers must wait in purgatory for a list admin to okay them  
via the mailman software.  Currently myself and a few other people  
monitor this and try and authorize posts as soon as possible.

There have been a few problems on the new mailing list and web server  
due mostly to it not having a lot of memory installed.  This has  
usually been the delay when the machine locked up and for us to do  
our remote-reset of the box.   You would have also noticed that the  
website was unreachable during that time.  It appears that Chris has  
installed the new mem yesterday so we hope this will be less of an  
issue in the future but we are still trying to investigate whether  
there is a memory leak in some of the software or if this was all due  
to the low-memory conditions.  This usual the problem with long  
mailing delays are a server problem and nothing particular that a  
user has to worry about.


-jason
On Apr 21, 2006, at 10:05 PM, Chris Fields wrote:

> This problem was mentioned by Daniel Bornman earlier today (the one  
> I sniped
> at for repeated posts).  He sent three emails which didn't make it  
> through
> until this afternoon.  I originally thought it might be the  
> changeover to
> the new server.  However, personally, I haven't seen any problems  
> today,
> though I have seen slowdowns before.  A post to the list went  
> through just
> fine this morning around the same time that Daniel sent an email to  
> the list
> (my post to the list was at 10:27 AM and I received it through the  
> list at
> 10:34 AM).  It could be a server problem on his end.  If your  
> running into
> the same problem it could be coincidence, but it likely is  
> something else.
> Bad timing?
>
> I also made CVS commits which showed up pretty quickly on bioperl- 
> guts-l
> yesterday.
>
> I believe Chris D. mentioned at some point that the mail to the  
> list goes
> through virus-checking and a spam-filters so that explains a slight  
> delay,
> but not several hours.
>
> Christopher Fields
> Postdoctoral Researcher - Switzer Lab
> Dept. of Biochemistry
> University of Illinois Urbana-Champaign
>
>
>> -----Original Message-----
>> From: bioperl-l-bounces at lists.open-bio.org [mailto:bioperl-l-
>> bounces at lists.open-bio.org] On Behalf Of Torsten Seemann
>> Sent: Friday, April 21, 2006 8:00 PM
>> To: bioperl-l at lists.open-bio.org
>> Subject: [Bioperl-l] Speed of bioperl mailing list distribution
>>
>> Hi,
>>
>> I've noticed that we often get repeated posts on the Bioperl lists,
>> as well as multiple duplicate responses that are hours apart in  
>> post time.
>>
>> I just thought I should mention that a post to bioperl-l (or an  
>> appearance
>> in
>> bioperl-guts-l after a CVS commit), from my point of view, can  
>> take up to
>> 10
>> hours to appear, at best 30 minutes. I'm not sure where the delay  
>> is but I
>> don't think it is at my end (?).
>>
>> So a new poster may post multiple times under the assumption that the
>> first
>> ones didn't work because the didn't appear in a timely fashion  
>> (for their
>> definition of timely :-). And responders will respond because they  
>> haven't
>> yet
>> received the previous response which already answered the post.
>>
>> Just thought I'd mention it - any one else have this problem?
>>
>> --
>> Torsten Seemann
>> Victorian Bioinformatics Consortium, Monash University, Australia
>> http://www.vicbioinformatics.com/
>> _______________________________________________
>> Bioperl-l mailing list
>> Bioperl-l at lists.open-bio.org
>> http://lists.open-bio.org/mailman/listinfo/bioperl-l
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/bioperl-l

--
Jason Stajich
Duke University
http://www.duke.edu/~jes12




More information about the Bioperl-l mailing list