Thread: Site way slow, please check
People, Starting at about 10pm PST, www.postgresql.org got *real* slow, according to several people in different countries. And we haven't been slashdotted yet! Please check on what could be slamming the servers. Thanks! -- --Josh Josh Berkus Aglio Database Solutions San Francisco
> -----Original Message----- > From: pgsql-www-owner@postgresql.org > [mailto:pgsql-www-owner@postgresql.org] On Behalf Of Josh Berkus > Sent: 19 January 2005 08:26 > To: pgsql-www@postgresql.org > Subject: [pgsql-www] Site way slow, please check > > People, > > Starting at about 10pm PST, www.postgresql.org got *real* > slow, according to > several people in different countries. And we haven't been > slashdotted yet! > > Please check on what could be slamming the servers. Thanks! Yup, I'm seeing that as well - judging by a quick ps, it *looks* like it's bittorrent that's killing it, along with a fair few hits on the ftp server. Unfortunately there's not much I can do as it's in a VM, and things like top don't work from there so I can't check properly. Going on IRC to chat about it I'm told you just went to sleep.... I'll do what I can, but I know nothing about BT :-( Regards, Dave.
Hi, Josh Berkus wrote: > Starting at about 10pm PST, www.postgresql.org got *real* slow, according to > several people in different countries. And we haven't been slashdotted yet! It *was* really slow a few hours ago, right now it looks belly-up, I'm getting timeouts. :[ Any backup plans? The annnouncements are already all over the Russian sites, I presume they'll soon appear everywhere.
I see redirect from www.postgresql.org to borg.postgresql.org. Weird.
Oleg
On Wed, 19 Jan 2005, Alexey Borzov wrote:
> Hi,
>
> Josh Berkus wrote:
>> Starting at about 10pm PST, www.postgresql.org got *real* slow, according
>> to several people in different countries. And we haven't been slashdotted
>> yet!
>
> It *was* really slow a few hours ago, right now it looks belly-up, I'm
> getting timeouts. :[
>
> Any backup plans? The annnouncements are already all over the Russian sites,
> I presume they'll soon appear everywhere.
>
> ---------------------------(end of broadcast)---------------------------
> TIP 8: explain analyze is your friend
>
Regards,
Oleg
_____________________________________________________________
Oleg Bartunov, sci.researcher, hostmaster of AstroNet,
Sternberg Astronomical Institute, Moscow University (Russia)
Internet: oleg@sai.msu.su, http://www.sai.msu.su/~megera/
phone: +007(095)939-16-83, +007(095)939-23-83
www.postgesql.org has been redirected by Dave to borg.postgresql.org to handle the load - www. fell over completely. Right now, we're waiting for Marc to answer Daves page and update the DNS record so we can get some round-robin between more than one server too - hopefully before we get hit with the actual slashdot effect when the US wakes up for real. //Magnus > -----Original Message----- > From: pgsql-www-owner@postgresql.org > [mailto:pgsql-www-owner@postgresql.org] On Behalf Of Oleg Bartunov > Sent: Wednesday, January 19, 2005 12:51 PM > To: Alexey Borzov > Cc: josh@agliodbs.com; pgsql-www@postgresql.org > Subject: Re: [pgsql-www] Site way slow, please check > > I see redirect from www.postgresql.org to borg.postgresql.org. Weird. > > Oleg > On Wed, 19 Jan 2005, Alexey Borzov wrote: > > > Hi, > > > > Josh Berkus wrote: > >> Starting at about 10pm PST, www.postgresql.org got *real* > slow, according > >> to several people in different countries. And we haven't > been slashdotted > >> yet! > > > > It *was* really slow a few hours ago, right now it looks > belly-up, I'm > > getting timeouts. :[ > > > > Any backup plans? The annnouncements are already all over > the Russian > > sites, I presume they'll soon appear everywhere. > > > > ---------------------------(end of > > broadcast)--------------------------- > > TIP 8: explain analyze is your friend > > > > Regards, > Oleg > _____________________________________________________________ > Oleg Bartunov, sci.researcher, hostmaster of AstroNet, > Sternberg Astronomical Institute, Moscow University (Russia) > Internet: oleg@sai.msu.su, http://www.sai.msu.su/~megera/ > phone: +007(095)939-16-83, +007(095)939-23-83 > > ---------------------------(end of > broadcast)--------------------------- > TIP 1: subscribe and unsubscribe commands go to > majordomo@postgresql.org >
On Wed, 19 Jan 2005, Magnus Hagander wrote: > www.postgesql.org has been redirected by Dave to borg.postgresql.org to > handle the load - www. fell over completely. I think redirecting just images would help ! It's very bad server which is mostly static couldn't hold load. It's interesting to see later what was peak rate. > > Right now, we're waiting for Marc to answer Daves page and update the > DNS record so we can get some round-robin between more than one server > too - hopefully before we get hit with the actual slashdot effect when > the US wakes up for real. > > //Magnus > >> -----Original Message----- >> From: pgsql-www-owner@postgresql.org >> [mailto:pgsql-www-owner@postgresql.org] On Behalf Of Oleg Bartunov >> Sent: Wednesday, January 19, 2005 12:51 PM >> To: Alexey Borzov >> Cc: josh@agliodbs.com; pgsql-www@postgresql.org >> Subject: Re: [pgsql-www] Site way slow, please check >> >> I see redirect from www.postgresql.org to borg.postgresql.org. Weird. >> >> Oleg >> On Wed, 19 Jan 2005, Alexey Borzov wrote: >> >>> Hi, >>> >>> Josh Berkus wrote: >>>> Starting at about 10pm PST, www.postgresql.org got *real* >> slow, according >>>> to several people in different countries. And we haven't >> been slashdotted >>>> yet! >>> >>> It *was* really slow a few hours ago, right now it looks >> belly-up, I'm >>> getting timeouts. :[ >>> >>> Any backup plans? The annnouncements are already all over >> the Russian >>> sites, I presume they'll soon appear everywhere. >>> >>> ---------------------------(end of >>> broadcast)--------------------------- >>> TIP 8: explain analyze is your friend >>> >> >> Regards, >> Oleg >> _____________________________________________________________ >> Oleg Bartunov, sci.researcher, hostmaster of AstroNet, >> Sternberg Astronomical Institute, Moscow University (Russia) >> Internet: oleg@sai.msu.su, http://www.sai.msu.su/~megera/ >> phone: +007(095)939-16-83, +007(095)939-23-83 >> >> ---------------------------(end of >> broadcast)--------------------------- >> TIP 1: subscribe and unsubscribe commands go to >> majordomo@postgresql.org >> > > ---------------------------(end of broadcast)--------------------------- > TIP 7: don't forget to increase your free space map settings > Regards, Oleg _____________________________________________________________ Oleg Bartunov, sci.researcher, hostmaster of AstroNet, Sternberg Astronomical Institute, Moscow University (Russia) Internet: oleg@sai.msu.su, http://www.sai.msu.su/~megera/ phone: +007(095)939-16-83, +007(095)939-23-83
> > www.postgesql.org has been redirected by Dave to > borg.postgresql.org > > to handle the load - www. fell over completely. > > I think redirecting just images would help ! It's very bad > server which is mostly static couldn't hold load. It's > interesting to see later what was peak rate. It fell over with less load than there is now, and right now borg is taking 18 req/sec. So it definitly fell at a very low load. Though the same box also runs the ftp server, and there was some load from that one. //Magnus
> -----Original Message----- > From: pgsql-www-owner@postgresql.org > [mailto:pgsql-www-owner@postgresql.org] On Behalf Of Magnus Hagander > Sent: 19 January 2005 12:13 > To: Oleg Bartunov > Cc: Alexey Borzov; josh@agliodbs.com; pgsql-www@postgresql.org > Subject: Re: [pgsql-www] Site way slow, please check > > > > www.postgesql.org has been redirected by Dave to > > borg.postgresql.org > > > to handle the load - www. fell over completely. > > > > I think redirecting just images would help ! It's very bad > > server which is mostly static couldn't hold load. It's > > interesting to see later what was peak rate. > > It fell over with less load than there is now, and right now borg is > taking 18 req/sec. So it definitly fell at a very low load. Though the > same box also runs the ftp server, and there was some load from that > one. And bittorrent, and rsync, and is one of the mail servers... Borg is otherwise idle. For interest: Current Time: Wednesday, 19-Jan-2005 04:29:29 PST Restart Time: Wednesday, 19-Jan-2005 03:39:46 PST Parent Server Generation: 7 Server uptime: 49 minutes 43 seconds Total accesses: 81525 - Total Traffic: 589.6 MB CPU Usage: u1.85156 s2.42188 cu0 cs0 - .143% CPU load 27.3 requests/sec - 202.4 kB/second - 7.4 kB/request 147 requests currently being processed, 83 idle servers /D
On Wed, 19 Jan 2005, Dave Page wrote:
>
>
>> -----Original Message-----
>> From: pgsql-www-owner@postgresql.org
>> [mailto:pgsql-www-owner@postgresql.org] On Behalf Of Magnus Hagander
>> Sent: 19 January 2005 12:13
>> To: Oleg Bartunov
>> Cc: Alexey Borzov; josh@agliodbs.com; pgsql-www@postgresql.org
>> Subject: Re: [pgsql-www] Site way slow, please check
>>
>>>> www.postgesql.org has been redirected by Dave to
>>> borg.postgresql.org
>>>> to handle the load - www. fell over completely.
>>>
>>> I think redirecting just images would help ! It's very bad
>>> server which is mostly static couldn't hold load. It's
>>> interesting to see later what was peak rate.
>>
>> It fell over with less load than there is now, and right now borg is
>> taking 18 req/sec. So it definitly fell at a very low load. Though the
>> same box also runs the ftp server, and there was some load from that
>> one.
>
> And bittorrent, and rsync, and is one of the mail servers...
>
> Borg is otherwise idle. For interest:
>
> Current Time: Wednesday, 19-Jan-2005 04:29:29 PST
> Restart Time: Wednesday, 19-Jan-2005 03:39:46 PST
> Parent Server Generation: 7
> Server uptime: 49 minutes 43 seconds
> Total accesses: 81525 - Total Traffic: 589.6 MB
> CPU Usage: u1.85156 s2.42188 cu0 cs0 - .143% CPU load
> 27.3 requests/sec - 202.4 kB/second - 7.4 kB/request
> 147 requests currently being processed, 83 idle servers
rather irresponsibly to mess all services.
>
> /D
>
Regards,
Oleg
_____________________________________________________________
Oleg Bartunov, sci.researcher, hostmaster of AstroNet,
Sternberg Astronomical Institute, Moscow University (Russia)
Internet: oleg@sai.msu.su, http://www.sai.msu.su/~megera/
phone: +007(095)939-16-83, +007(095)939-23-83
> -----Original Message----- > From: Oleg Bartunov [mailto:oleg@sai.msu.su] > Sent: 19 January 2005 14:02 > To: Dave Page > Cc: Magnus Hagander; Alexey Borzov; josh@agliodbs.com; > pgsql-www@postgresql.org > Subject: RE: [pgsql-www] Site way slow, please check > > rather irresponsibly to mess all services. Huh? Sorry, don't understand that. Regards, Dave.
> -----Original Message----- > From: Oleg Bartunov [mailto:oleg@sai.msu.su] > Sent: 19 January 2005 14:40 > To: Dave Page > Cc: Magnus Hagander; Alexey Borzov; josh@agliodbs.com; > pgsql-www@postgresql.org > Subject: RE: [pgsql-www] Site way slow, please check > > On Wed, 19 Jan 2005, Dave Page wrote: > > > > > > >> -----Original Message----- > >> From: Oleg Bartunov [mailto:oleg@sai.msu.su] > >> Sent: 19 January 2005 14:02 > >> To: Dave Page > >> Cc: Magnus Hagander; Alexey Borzov; josh@agliodbs.com; > >> pgsql-www@postgresql.org > >> Subject: RE: [pgsql-www] Site way slow, please check > >> > >> rather irresponsibly to mess all services. > > > > Huh? Sorry, don't understand that. > > I meant, predicting "warmed up" interest to > www.postgresql.org and knowing > about not well-scaled web architecture, messing all services > on one server > was not a wise solution. We're lucky, that, you, Dave and > Magnus were in place. > Next time, lets think before. I proposed several times > architecture that > works well in very big servers. Nothing new, it's already de > facto standard. We have thought before, in fact I for one have done a *lot* of work on the mirroring and related bits of the infrastructure. The problem is one of time - I simply didn't have enough to get all my planned work done before release. BTW, the problems seen today seem to be entirely caused by Apache's access logging. With it turned off, the 3 servers hosting www.postgresql.org are purring along (including svr4 which now has rsync re-enabled). /D
On Wed, 19 Jan 2005, Dave Page wrote:
>
>
>> -----Original Message-----
>> From: Oleg Bartunov [mailto:oleg@sai.msu.su]
>> Sent: 19 January 2005 14:02
>> To: Dave Page
>> Cc: Magnus Hagander; Alexey Borzov; josh@agliodbs.com;
>> pgsql-www@postgresql.org
>> Subject: RE: [pgsql-www] Site way slow, please check
>>
>> rather irresponsibly to mess all services.
>
> Huh? Sorry, don't understand that.
I meant, predicting "warmed up" interest to www.postgresql.org and knowing
about not well-scaled web architecture, messing all services on one server
was not a wise solution. We're lucky, that, you, Dave and Magnus were in place.
Next time, lets think before. I proposed several times architecture that
works well in very big servers. Nothing new, it's already de facto standard.
>
> Regards, Dave.
>
Regards,
Oleg
_____________________________________________________________
Oleg Bartunov, sci.researcher, hostmaster of AstroNet,
Sternberg Astronomical Institute, Moscow University (Russia)
Internet: oleg@sai.msu.su, http://www.sai.msu.su/~megera/
phone: +007(095)939-16-83, +007(095)939-23-83
On Wed, 19 Jan 2005, Dave Page wrote:
>
>
>> -----Original Message-----
>> From: Oleg Bartunov [mailto:oleg@sai.msu.su]
>> Sent: 19 January 2005 14:40
>> To: Dave Page
>> Cc: Magnus Hagander; Alexey Borzov; josh@agliodbs.com;
>> pgsql-www@postgresql.org
>> Subject: RE: [pgsql-www] Site way slow, please check
>>
>> On Wed, 19 Jan 2005, Dave Page wrote:
>>
>>>
>>>
>>>> -----Original Message-----
>>>> From: Oleg Bartunov [mailto:oleg@sai.msu.su]
>>>> Sent: 19 January 2005 14:02
>>>> To: Dave Page
>>>> Cc: Magnus Hagander; Alexey Borzov; josh@agliodbs.com;
>>>> pgsql-www@postgresql.org
>>>> Subject: RE: [pgsql-www] Site way slow, please check
>>>>
>>>> rather irresponsibly to mess all services.
>>>
>>> Huh? Sorry, don't understand that.
>>
>> I meant, predicting "warmed up" interest to
>> www.postgresql.org and knowing
>> about not well-scaled web architecture, messing all services
>> on one server
>> was not a wise solution. We're lucky, that, you, Dave and
>> Magnus were in place.
>> Next time, lets think before. I proposed several times
>> architecture that
>> works well in very big servers. Nothing new, it's already de
>> facto standard.
>
> We have thought before, in fact I for one have done a *lot* of work on
> the mirroring and related bits of the infrastructure. The problem is one
> of time - I simply didn't have enough to get all my planned work done
> before release.
I see, let me know when you decide to work on infrastructure.
>
> BTW, the problems seen today seem to be entirely caused by Apache's
> access logging. With it turned off, the 3 servers hosting
> www.postgresql.org are purring along (including svr4 which now has rsync
> re-enabled).
dns-resolving in apache logs was switched off and redirected logs to separate
disk ?
>
> /D
>
Regards,
Oleg
_____________________________________________________________
Oleg Bartunov, sci.researcher, hostmaster of AstroNet,
Sternberg Astronomical Institute, Moscow University (Russia)
Internet: oleg@sai.msu.su, http://www.sai.msu.su/~megera/
phone: +007(095)939-16-83, +007(095)939-23-83
> -----Original Message----- > From: Oleg Bartunov [mailto:oleg@sai.msu.su] > Sent: 19 January 2005 16:06 > To: Dave Page > Cc: Magnus Hagander; Alexey Borzov; josh@agliodbs.com; > pgsql-www@postgresql.org > Subject: RE: [pgsql-www] Site way slow, please check > > > > > BTW, the problems seen today seem to be entirely caused by Apache's > > access logging. With it turned off, the 3 servers hosting > > www.postgresql.org are purring along (including svr4 which > now has rsync > > re-enabled). > > dns-resolving in apache logs was switched off and redirected > logs to separate > disk ? Logs were switched off altogether (the servers only have one raid array). Regards, Dave
On Wed, 19 Jan 2005, Oleg Bartunov wrote: >> BTW, the problems seen today seem to be entirely caused by Apache's >> access logging. With it turned off, the 3 servers hosting >> www.postgresql.org are purring along (including svr4 which now has rsync >> re-enabled). > > dns-resolving in apache logs was switched off and redirected logs to > separate disk ? dns resolving has never been turned on, but Dave switched AccessLog to point to /dev/null ... ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664
>> Please check on what could be slamming the servers. Thanks! > > Yup, I'm seeing that as well - judging by a quick ps, it *looks* like > it's bittorrent that's killing it, along with a fair few hits on the ftp > server. Unfortunately there's not much I can do as it's in a VM, and > things like top don't work from there so I can't check properly. > > Going on IRC to chat about it I'm told you just went to sleep.... I'll > do what I can, but I know nothing about BT :-( For what it's worth (and I'm on the digest list so someone may have already brought this up), bittorrent is fairly resource intensive and perhaps it doesn't need to be run on this particular server. I setup bt seeds for all the torrents on a box I had last night with a 10 megabit line, and I'll do the same with my other production box today. As long as there's at least one torrent seed going out there, I don't think the webserver should be seeding torrents.
> > >> Please check on what could be slamming the servers. Thanks! > > > > Yup, I'm seeing that as well - judging by a quick ps, it > *looks* like > > it's bittorrent that's killing it, along with a fair few > hits on the > > ftp server. Unfortunately there's not much I can do as it's > in a VM, > > and things like top don't work from there so I can't check properly. > > > > Going on IRC to chat about it I'm told you just went to > sleep.... I'll > > do what I can, but I know nothing about BT :-( > > For what it's worth (and I'm on the digest list so someone > may have already brought this up), bittorrent is fairly > resource intensive and perhaps it doesn't need to be run on > this particular server. I setup bt seeds for all the > torrents on a box I had last night with a 10 megabit line, > and I'll do the same with my other production box today. As > long as there's at least one torrent seed going out there, I > don't think the webserver should be seeding torrents. There have been several other machines seeding. Only the trackers are on that box. I think. At least there were plenty of other seeders available on different networks. //Magnus
On Wed, 19 Jan 2005, Magnus Hagander wrote: >> >>>> Please check on what could be slamming the servers. Thanks! >>> >>> Yup, I'm seeing that as well - judging by a quick ps, it >> *looks* like >>> it's bittorrent that's killing it, along with a fair few >> hits on the >>> ftp server. Unfortunately there's not much I can do as it's >> in a VM, >>> and things like top don't work from there so I can't check properly. >>> >>> Going on IRC to chat about it I'm told you just went to >> sleep.... I'll >>> do what I can, but I know nothing about BT :-( >> >> For what it's worth (and I'm on the digest list so someone >> may have already brought this up), bittorrent is fairly >> resource intensive and perhaps it doesn't need to be run on >> this particular server. I setup bt seeds for all the >> torrents on a box I had last night with a 10 megabit line, >> and I'll do the same with my other production box today. As >> long as there's at least one torrent seed going out there, I >> don't think the webserver should be seeding torrents. > > There have been several other machines seeding. Only the trackers are on > that box. I think. At least there were plenty of other seeders available > on different networks. As far as I know, this is the case ... I've been watching that server all day, and all appears to be running smoothly ... ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664
> -----Original Message----- > From: Marc G. Fournier [mailto:scrappy@postgresql.org] > Sent: 19 January 2005 19:42 > To: Magnus Hagander > Cc: fred@redhotpenguin.com; Dave Page; josh@agliodbs.com; > pgsql-www@postgresql.org > Subject: Re: [pgsql-www] Site way slow, please check > > >> For what it's worth (and I'm on the digest list so someone > >> may have already brought this up), bittorrent is fairly > >> resource intensive and perhaps it doesn't need to be run on > >> this particular server. I setup bt seeds for all the > >> torrents on a box I had last night with a 10 megabit line, > >> and I'll do the same with my other production box today. As > >> long as there's at least one torrent seed going out there, I > >> don't think the webserver should be seeding torrents. > > > > There have been several other machines seeding. Only the > trackers are on > > that box. I think. At least there were plenty of other > seeders available > > on different networks. > > As far as I know, this is the case ... I've been watching > that server all > day, and all appears to be running smoothly ... We do seem to have lots of very old versions in there as well though - is there any need for that? I would think that 8.0, 7.4.6 and any other popular downloads (psqlodbc, pgadmin etc) would have been enough? Regards, Dave
>>As far as I know, this is the case ... I've been watching >>that server all >>day, and all appears to be running smoothly ... >> >> > >We do seem to have lots of very old versions in there as well though - >is there any need for that? I would think that 8.0, 7.4.6 and any other >popular downloads (psqlodbc, pgadmin etc) would have been enough? > > 7.3.8 as well SIncerely, Joshua D. Drkae >Regards, Dave > >---------------------------(end of broadcast)--------------------------- >TIP 4: Don't 'kill -9' the postmaster > > -- Command Prompt, Inc., home of Mammoth PostgreSQL - S/ODBC and S/JDBC Postgresql support, programming shared hosting and dedicated hosting. +1-503-667-4564 - jd@commandprompt.com - http://www.commandprompt.com PostgreSQL Replicator -- production quality replication for PostgreSQL
Attachment
On Wed, 19 Jan 2005, Joshua D. Drake wrote: > >>> As far as I know, this is the case ... I've been watching that server all >>> day, and all appears to be running smoothly ... >>> >> >> We do seem to have lots of very old versions in there as well though - >> is there any need for that? I would think that 8.0, 7.4.6 and any other >> popular downloads (psqlodbc, pgadmin etc) would have been enough? >> > 7.3.8 as well Sorry, am confusd ... what are y'all looking at? Oh hadn't noticed the subdirectories before ... I know several ppl still stuck in 7.3 land, and with stuff like Slony requiring a source download, having it there, I would think, shouldn't be a problem ... ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664
> -----Original Message----- > From: Marc G. Fournier [mailto:scrappy@postgresql.org] > Sent: 20 January 2005 01:24 > To: Joshua D. Drake > Cc: Dave Page; Marc G. Fournier; Magnus Hagander; > fred@redhotpenguin.com; josh@agliodbs.com; pgsql-www@postgresql.org > Subject: Re: [pgsql-www] Site way slow, please check > > Sorry, am confusd ... what are y'all looking at? Oh hadn't > noticed the > subdirectories before ... I know several ppl still stuck in > 7.3 land, and > with stuff like Slony requiring a source download, having it there, I > would think, shouldn't be a problem ... Hmm, looking more closely, it seems there's only a few things that could probably go - 7.2.x, CVSUP and NT_Support_Files for example. However, how much load does this take (I got varying reports from none to loads on IRC)? Perhaps we should scrounge another server and offload all the bt stuff? Regards, Dave
> > >> -----Original Message----- >> From: Marc G. Fournier [mailto:scrappy@postgresql.org] >> Sent: 20 January 2005 01:24 >> To: Joshua D. Drake >> Cc: Dave Page; Marc G. Fournier; Magnus Hagander; >> fred@redhotpenguin.com; josh@agliodbs.com; pgsql-www@postgresql.org >> Subject: Re: [pgsql-www] Site way slow, please check >> >> Sorry, am confusd ... what are y'all looking at? Oh hadn't >> noticed the >> subdirectories before ... I know several ppl still stuck in >> 7.3 land, and >> with stuff like Slony requiring a source download, having it there, I >> would think, shouldn't be a problem ... > > Hmm, looking more closely, it seems there's only a few things that could > probably go - 7.2.x, CVSUP and NT_Support_Files for example. > > However, how much load does this take (I got varying reports from none > to loads on IRC)? Perhaps we should scrounge another server and offload > all the bt stuff? You are probably running a tracker, I haven't run one but I do know about the seeding resource usage. Seeders use one bittorrent process per seed, and those are the machines where bt puts load on. I think we should bt every single release and list it on that bt page.. It's easier to maintain, and the seeders can start torrent seeds according to their resources.
> -----Original Message----- > From: Fred Moyer [mailto:fred@redhotpenguin.com] > Sent: 20 January 2005 09:30 > To: Dave Page > Cc: Marc G. Fournier; Joshua D. Drake; Magnus Hagander; > fred@redhotpenguin.com; josh@agliodbs.com; pgsql-www@postgresql.org > Subject: RE: [pgsql-www] Site way slow, please check > > > I think we should bt every single release and list it on that > bt page.. > It's easier to maintain, and the seeders can start torrent > seeds according > to their resources. Actually, Magnus had a great idea yesterday - he proprosed that the FTP browser (www.postgresql.org/ftp/) should have a [BitTorrent] link on each file, allowing the same user interface to be used. Of course, that means publishing every file in ftp, rather than fewer, but if that can be done automatically, and on a dedicated server it might work well. Regards, Dave.