Обсуждение: New email list for emergency communications
[ BCC to core.] Right now we have no way to communicate as a group if our community email infrastructure isn't working. Therefore, I have created an email list on my machine that can be used for communication in such situations. It is called: pgsql-emergency@momjian.us I envision this as an email list that will allow more rapid recovery during email outages. I know many use IRC in such cases but this allows notification. Also some don't use IRC. Anyone can post to the email list. There is no moderation, though the list sits behind my spam filters. The list will not be archived. If you want to be subscribed, please email me privately. I think it is particularly important for infrastructure managers to be subscribed to this list. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 24 Mar 2008 11:17:02 -0400 (EDT) Bruce Momjian <bruce@momjian.us> wrote: > > [ BCC to core.] > > Right now we have no way to communicate as a group if our community > email infrastructure isn't working. Yes there is. jabber.postgresql.org irc.freenode.net direct email Not to mention how many other services people are using with buddy lists? :P > > Therefore, I have created an email list on my machine that can be used > for communication in such situations. It is called: > > pgsql-emergency@momjian.us > > I envision this as an email list that will allow more rapid recovery > during email outages. I know many use IRC in such cases but this > allows notification. Also some don't use IRC. > Everyone in the know uses one of: gtalk yahoo irc jabber > If you want to be subscribed, please email me privately. I think it > is particularly important for infrastructure managers to be > subscribed to this list. I would like the requirements defined by the -www and sysadmins teams. I don't want us to go off and create yet more infrastructure without full documentation and understanding of the purpose and whether it is redundant. This list is currently redundant as we have the resources and technology already in place to deal with this. I think you are attacking the wrong problem Bruce. Sincerely, Joshua D. Drake - -- The PostgreSQL Company since 1997: http://www.commandprompt.com/ PostgreSQL Community Conference: http://www.postgresqlconference.org/ United States PostgreSQL Association: http://www.postgresql.us/ Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFH58o1ATb/zqfZUUQRAoz6AJ9NQaM6oKuu41Lkc/iSqWTg291HIQCeOPm2 mG3hTV8NPKGr5TM48m19HA8= =OOnX -----END PGP SIGNATURE-----
Joshua D. Drake wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On Mon, 24 Mar 2008 11:17:02 -0400 (EDT) > Bruce Momjian <bruce@momjian.us> wrote: > > > > > [ BCC to core.] > > > > Right now we have no way to communicate as a group if our community > > email infrastructure isn't working. > > Yes there is. > > jabber.postgresql.org > irc.freenode.net > direct email I think we need something more centralized, and with chat you have to know to look there for problems. I think we need something that allows people to notify us if there is a problem. > > If you want to be subscribed, please email me privately. I think it > > is particularly important for infrastructure managers to be > > subscribed to this list. > > I would like the requirements defined by the -www and sysadmins teams. > I don't want us to go off and create yet more infrastructure without > full documentation and understanding of the purpose and whether it is > redundant. > > This list is currently redundant as we have the resources and > technology already in place to deal with this. I think you are > attacking the wrong problem Bruce. Again, see above. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
On Mon, 24 Mar 2008, Joshua D. Drake wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On Mon, 24 Mar 2008 11:17:02 -0400 (EDT) > Bruce Momjian <bruce@momjian.us> wrote: > >> >> [ BCC to core.] >> >> Right now we have no way to communicate as a group if our community >> email infrastructure isn't working. > > Yes there is. > > jabber.postgresql.org > irc.freenode.net > direct email > > Not to mention how many other services people are using with buddy > lists? :P > >> >> Therefore, I have created an email list on my machine that can be used >> for communication in such situations. It is called: >> >> pgsql-emergency@momjian.us >> >> I envision this as an email list that will allow more rapid recovery >> during email outages. I know many use IRC in such cases but this >> allows notification. Also some don't use IRC. >> > > Everyone in the know uses one of: > > gtalk > yahoo > irc > jabber > >> If you want to be subscribed, please email me privately. I think it >> is particularly important for infrastructure managers to be >> subscribed to this list. > > I would like the requirements defined by the -www and sysadmins teams. > I don't want us to go off and create yet more infrastructure without > full documentation and understanding of the purpose and whether it is > redundant. > > This list is currently redundant as we have the resources and > technology already in place to deal with this. I think you are > attacking the wrong problem Bruce. For how often it happens, its worth nothing: I agree 100% with Joshua here, but, if Bruce wants to create a list on his personal domain (which, in itself is a failure point as it requires that *his* domain is not 100% perfect), it doesn't mean that anyone actually has to *use* it ... I know I never will, since there already exist faster, more reliable, contact methods ...
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 24 Mar 2008 11:36:55 -0400 (EDT) Bruce Momjian <bruce@momjian.us> wrote: > I think we need something more centralized, and with chat you have to > know to look there for problems. I think we need something that > allows people to notify us if there is a problem. We are attacking the wrong problem. 99% of the time we know when there is a problem. I leave the 1% only because we aren't perfect. Sincerely, Joshua D. Drake - -- The PostgreSQL Company since 1997: http://www.commandprompt.com/ PostgreSQL Community Conference: http://www.postgresqlconference.org/ United States PostgreSQL Association: http://www.postgresql.us/ Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFH58zQATb/zqfZUUQRAiNHAJsHC8F0Y6l78/I469zXzPUbDxGKsACgraN1 b5pRzEUiX96s1qnMDFFnLmc= =BBKP -----END PGP SIGNATURE-----
Marc G. Fournier wrote: > > I would like the requirements defined by the -www and sysadmins teams. > > I don't want us to go off and create yet more infrastructure without > > full documentation and understanding of the purpose and whether it is > > redundant. > > > > This list is currently redundant as we have the resources and > > technology already in place to deal with this. I think you are > > attacking the wrong problem Bruce. > > For how often it happens, its worth nothing: I agree 100% with Joshua > here, but, if Bruce wants to create a list on his personal domain (which, > in itself is a failure point as it requires that *his* domain is not 100% > perfect), it doesn't mean that anyone actually has to *use* it ... I know > I never will, since there already exist faster, more reliable, contact > methods ... I think we need a more centralized contact system than we have now. Personally when I don't get email I am unsure who to contact (because community email is down). The email list was a method to centralize that. The list of possible contact methods listed confirms my point. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Joshua D. Drake wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On Mon, 24 Mar 2008 11:36:55 -0400 (EDT) > Bruce Momjian <bruce@momjian.us> wrote: > > > > I think we need something more centralized, and with chat you have to > > know to look there for problems. I think we need something that > > allows people to notify us if there is a problem. > > We are attacking the wrong problem. 99% of the time we know when > there is a problem. I leave the 1% only because we aren't perfect. Well, I certainly don't until I realize I haven't received email, nor do I know who is working on it. Right now the emergency communication system is ad-hock. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Bruce Momjian wrote: > I think we need a more centralized contact system than we have now. > Personally when I don't get email I am unsure who to contact (because > community email is down). The email list was a method to centralize > that. The list of possible contact methods listed confirms my point. Perhaps you should have a jabber.postgresql.org account? -- Alvaro Herrera http://www.CommandPrompt.com/ The PostgreSQL Company - Command Prompt, Inc.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 24 Mar 2008 11:47:20 -0400 (EDT) Bruce Momjian <bruce@momjian.us> wrote: > > For how often it happens, its worth nothing: I agree 100% with > > Joshua here, but, if Bruce wants to create a list on his personal > > domain (which, in itself is a failure point as it requires that > > *his* domain is not 100% perfect), it doesn't mean that anyone > > actually has to *use* it ... I know I never will, since there > > already exist faster, more reliable, contact methods ... > > I think we need a more centralized contact system than we have now. > Personally when I don't get email I am unsure who to contact (because > community email is down). The email list was a method to centralize > that. The list of possible contact methods listed confirms my point. This comes back to defining teams which is already being discussed in another thread. The people that need to be contacted are the sysadmins team but we need to know how. Right now, the only people that know how to get a hold of the sysadmins team, is the sysadmins team :P. As I said previously, we are attacking the wrong problem. The problem is not detection (although it could be improved), the problem is not communication (see my 99% remark). The problem is one of complexity of environment and resources to manage that environment. This is already being addressed on the sysadmins list and I expect that it will be resolved within 2 weeks. Does that sound about right Marc? Sincerely, Joshua D. Drake - -- The PostgreSQL Company since 1997: http://www.commandprompt.com/ PostgreSQL Community Conference: http://www.postgresqlconference.org/ United States PostgreSQL Association: http://www.postgresql.us/ Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFH5852ATb/zqfZUUQRAkJ5AJ9iJF/nz4NmRxztsOzfthc1a3BzYgCeJrHo chHIksxGEqxbBCPCosPVzYc= =fLGa -----END PGP SIGNATURE-----
Alvaro Herrera wrote: > Bruce Momjian wrote: > > > I think we need a more centralized contact system than we have now. > > Personally when I don't get email I am unsure who to contact (because > > community email is down). The email list was a method to centralize > > that. The list of possible contact methods listed confirms my point. > > Perhaps you should have a jabber.postgresql.org account? What does that do for me that my existing jabber account does not? -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 24 Mar 2008 12:51:19 -0300 Alvaro Herrera <alvherre@commandprompt.com> wrote: > Bruce Momjian wrote: > > > I think we need a more centralized contact system than we have now. > > Personally when I don't get email I am unsure who to contact > > (because community email is down). The email list was a method to > > centralize that. The list of possible contact methods listed > > confirms my point. > > Perhaps you should have a jabber.postgresql.org account? > That seems reasonable. Sincerely, Joshua D. Drake - -- The PostgreSQL Company since 1997: http://www.commandprompt.com/ PostgreSQL Community Conference: http://www.postgresqlconference.org/ United States PostgreSQL Association: http://www.postgresql.us/ Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFH586aATb/zqfZUUQRArDaAJ9rXQCgmzYAsUAGvwAGxUXf1j6RkACeMOqe wKHIZGtgACT1speiUq9JuYc= =zBmY -----END PGP SIGNATURE-----
Joshua D. Drake wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On Mon, 24 Mar 2008 11:47:20 -0400 (EDT) > Bruce Momjian <bruce@momjian.us> wrote: > > > > For how often it happens, its worth nothing: I agree 100% with > > > Joshua here, but, if Bruce wants to create a list on his personal > > > domain (which, in itself is a failure point as it requires that > > > *his* domain is not 100% perfect), it doesn't mean that anyone > > > actually has to *use* it ... I know I never will, since there > > > already exist faster, more reliable, contact methods ... > > > > I think we need a more centralized contact system than we have now. > > Personally when I don't get email I am unsure who to contact (because > > community email is down). The email list was a method to centralize > > that. The list of possible contact methods listed confirms my point. > > This comes back to defining teams which is already being discussed in > another thread. The people that need to be contacted are the sysadmins > team but we need to know how. Right now, the only people that know how > to get a hold of the sysadmins team, is the sysadmins team :P. My big point is if the sysadmins team is hosted by the postgresql.org domain, how do we communicate with them when postgresql.org email is down? -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 24 Mar 2008 11:52:25 -0400 (EDT) Bruce Momjian <bruce@momjian.us> wrote: > > Perhaps you should have a jabber.postgresql.org account? > > What does that do for me that my existing jabber account does not? > Your jabber account can not talk to jdrake@jabber.postgresql.org when there is a problem with archives. Marc should really be on it too but besides you too, all of the sysadmins and -www team is already on it. Sincerely, Joshua D. Drake - -- The PostgreSQL Company since 1997: http://www.commandprompt.com/ PostgreSQL Community Conference: http://www.postgresqlconference.org/ United States PostgreSQL Association: http://www.postgresql.us/ Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFH589aATb/zqfZUUQRAhjcAJ9jqprH3v6IenXP4oObgeRTDGzqowCfeZOj gJlXPyr5XUBb5aOYpRkiWbU= =pVAM -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 24 Mar 2008 11:54:38 -0400 (EDT) Bruce Momjian <bruce@momjian.us> wrote: > > This comes back to defining teams which is already being discussed > > in another thread. The people that need to be contacted are the > > sysadmins team but we need to know how. Right now, the only people > > that know how to get a hold of the sysadmins team, is the sysadmins > > team :P. > > My big point is if the sysadmins team is hosted by the postgresql.org > domain, how do we communicate with them when postgresql.org email is > down? > See point about jabber. Joshua D. Drake - -- The PostgreSQL Company since 1997: http://www.commandprompt.com/ PostgreSQL Community Conference: http://www.postgresqlconference.org/ United States PostgreSQL Association: http://www.postgresql.us/ Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFH589wATb/zqfZUUQRAl9gAKCKs+TC05CVCr1mMPSK0GyeDYljigCglW7S QsqwR4zhKzIPabvW+WKdkhk= =t8hU -----END PGP SIGNATURE-----
Joshua D. Drake wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On Mon, 24 Mar 2008 11:52:25 -0400 (EDT) > Bruce Momjian <bruce@momjian.us> wrote: > > > > Perhaps you should have a jabber.postgresql.org account? > > > > What does that do for me that my existing jabber account does not? > > > > Your jabber account can not talk to jdrake@jabber.postgresql.org when > there is a problem with archives. Marc should really be on it too but > besides you too, all of the sysadmins and -www team is already on it. OK, sign me up (time for another Pidgin account). -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Bruce Momjian wrote: > [ BCC to core.] > > Right now we have no way to communicate as a group if our community > email infrastructure isn't working. > > Therefore, I have created an email list on my machine that can be used > for communication in such situations. It is called: > > pgsql-emergency@momjian.us > > I envision this as an email list that will allow more rapid recovery > during email outages. I know many use IRC in such cases but this allows > notification. Also some don't use IRC. > > Anyone can post to the email list. There is no moderation, though the > list sits behind my spam filters. The list will not be archived. > > If you want to be subscribed, please email me privately. I think it is > particularly important for infrastructure managers to be subscribed to > this list. > There is one for sysadmins already. Every email on sysadmins has "If the PostgreSQL.org mailing lists are down, use the auxillary list pgsysadmins@agliodbs.com " I'm nto sure if that one is synced automatically though. Josh? //Magnus
On Mon, Mar 24, 2008 at 3:17 PM, Bruce Momjian <bruce@momjian.us> wrote: > > [ BCC to core.] > > Right now we have no way to communicate as a group if our community > email infrastructure isn't working. > > Therefore, I have created an email list on my machine that can be used > for communication in such situations. It is called: > > pgsql-emergency@momjian.us Thanks Bruce, but we've had such a list for a couple of years already. -- Dave Page EnterpriseDB UK Ltd: http://www.enterprisedb.com PostgreSQL UK 2008 Conference: http://www.postgresql.org.uk
Bruce Momjian wrote: > Marc G. Fournier wrote: >>> I would like the requirements defined by the -www and sysadmins teams. >>> I don't want us to go off and create yet more infrastructure without >>> full documentation and understanding of the purpose and whether it is >>> redundant. >>> >>> This list is currently redundant as we have the resources and >>> technology already in place to deal with this. I think you are >>> attacking the wrong problem Bruce. >> For how often it happens, its worth nothing: I agree 100% with Joshua >> here, but, if Bruce wants to create a list on his personal domain (which, >> in itself is a failure point as it requires that *his* domain is not 100% >> perfect), it doesn't mean that anyone actually has to *use* it ... I know >> I never will, since there already exist faster, more reliable, contact >> methods ... > > I think we need a more centralized contact system than we have now. > Personally when I don't get email I am unsure who to contact (because > community email is down). The email list was a method to centralize > that. The list of possible contact methods listed confirms my point. the sysadmin team already maintains a list of emergency contacts including external email, phone numbers, alternative contacts in addition to the what we are usually using to communicate (IM,personal email, the lists) - not sure if we need to duplicate that ... Stefan
Dave Page wrote: > On Mon, Mar 24, 2008 at 3:17 PM, Bruce Momjian <bruce@momjian.us> wrote: > > > > [ BCC to core.] > > > > Right now we have no way to communicate as a group if our community > > email infrastructure isn't working. > > > > Therefore, I have created an email list on my machine that can be used > > for communication in such situations. It is called: > > > > pgsql-emergency@momjian.us > > Thanks Bruce, but we've had such a list for a couple of years already. Oh, I didn't know. Email list removed. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
On Mon, Mar 24, 2008 at 11:48:36AM -0400, Bruce Momjian wrote: > Well, I certainly don't until I realize I haven't received email, nor do > I know who is working on it. Right now the emergency communication > system is ad-hock. So the solution you propose does not solve the problem you think you have -- i.e. that in the event there is a failure, the community doesn't know it. Would a health check system that updated web pages based on the monitor status help? (Hint: I think the monitors are already in place. What we need is some additional ways of accessing that data, I guess.) A
On Mon, Mar 24, 2008 at 11:57:39AM -0400, Bruce Momjian wrote: > > besides you too, all of the sysadmins and -www team is already on it. > > OK, sign me up (time for another Pidgin account). You shouldn't need more than one account -- jabber should allow authentication against any jabber server. A
Dave, >> pgsql-emergency@momjian.us > > Thanks Bruce, but we've had such a list for a couple of years already. > Where is that list? I can't remember the name of it, and it's had no traffic ... --Josh
Andrew Sullivan wrote: > On Mon, Mar 24, 2008 at 11:48:36AM -0400, Bruce Momjian wrote: > > Well, I certainly don't until I realize I haven't received email, nor do > > I know who is working on it. Right now the emergency communication > > system is ad-hock. > > So the solution you propose does not solve the problem you think you have > -- i.e. that in the event there is a failure, the community doesn't know it. > > Would a health check system that updated web pages based on the monitor > status help? (Hint: I think the monitors are already in place. What we > need is some additional ways of accessing that data, I guess.) Personally, I would love to get an email when the community email system isn't working, and status while it is being worked on. I assume others would like to as well, so I don't go reporting/contacting people when things are already being worked on. Right now, when email is down, I just start IM'ing people. A more structured system would help me. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Andrew Sullivan wrote: > On Mon, Mar 24, 2008 at 11:57:39AM -0400, Bruce Momjian wrote: > > > besides you too, all of the sysadmins and -www team is already on it. > > > > OK, sign me up (time for another Pidgin account). > > You shouldn't need more than one account -- jabber should allow > authentication against any jabber server. I am told our jabber system is private. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Josh Berkus wrote: > Dave, > > >> pgsql-emergency@momjian.us > > > > Thanks Bruce, but we've had such a list for a couple of years already. > > > > Where is that list? I can't remember the name of it, and it's had no > traffic ... Read later in the thread. It is at agliodbs.com (your domain). -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
On Mon, Mar 24, 2008 at 01:39:07PM -0400, Bruce Momjian wrote: > I am told our jabber system is private. That's a pity. It dramatically reduces the utility of the system, and I've yet to hear a justification for it in any installation I've seen. A
On Mon, Mar 24, 2008 at 5:37 PM, Josh Berkus <josh@agliodbs.com> wrote: > Dave, > > >> pgsql-emergency@momjian.us > > > > Thanks Bruce, but we've had such a list for a couple of years already. > > > > Where is that list? I can't remember the name of it, and it's had no > traffic ... pgsysadmins@agliodbs.com And no, I don't think it's ever been used because in the event of major problems Magnus, Stefan and I communicate via MSN, Jabber, email or phone - and when Marc is needed and is asleep (which tends to be the case at the time when things generally seem to go wrong), we page him. -- Dave Page EnterpriseDB UK Ltd: http://www.enterprisedb.com PostgreSQL UK 2008 Conference: http://www.postgresql.org.uk
Andrew Sullivan wrote: > On Mon, Mar 24, 2008 at 01:39:07PM -0400, Bruce Momjian wrote: > > I am told our jabber system is private. > > That's a pity. It dramatically reduces the utility of the system, and > I've yet to hear a justification for it in any installation I've seen. Agreed. It would be very nice if it were a usable system for the outside world. -- Alvaro Herrera http://www.CommandPrompt.com/ The PostgreSQL Company - Command Prompt, Inc.
On Mon, 24 Mar 2008 14:54:18 -0300 Alvaro Herrera wrote: > Andrew Sullivan wrote: > > On Mon, Mar 24, 2008 at 01:39:07PM -0400, Bruce Momjian wrote: > > > I am told our jabber system is private. > > > > That's a pity. It dramatically reduces the utility of the system, and > > I've yet to hear a justification for it in any installation I've seen. > > Agreed. It would be very nice if it were a usable system for the > outside world. Yes, i was asked more than once by ppl using a Google talk account. Would be nice to have them added without adding another account in pidgin. Kind regards -- Andreas 'ads' Scherbaum German PostgreSQL User Group
Bruce Momjian wrote: > Andrew Sullivan wrote: >> On Mon, Mar 24, 2008 at 11:48:36AM -0400, Bruce Momjian wrote: >>> Well, I certainly don't until I realize I haven't received email, nor do >>> I know who is working on it. Right now the emergency communication >>> system is ad-hock. >> So the solution you propose does not solve the problem you think you have >> -- i.e. that in the event there is a failure, the community doesn't know it. >> >> Would a health check system that updated web pages based on the monitor >> status help? (Hint: I think the monitors are already in place. What we >> need is some additional ways of accessing that data, I guess.) > > Personally, I would love to get an email when the community email system > isn't working, and status while it is being worked on. I assume others > would like to as well, so I don't go reporting/contacting people when > things are already being worked on. there is a large grey area here and things are by far not as black & white as your are painting them - like just defining what constitutes "community email not working" is not easy - he have a few dozends checks on that already and you can assume that we are aware of issues usually faster than most other people. For the reporting/contacting stuff it seems you actually want to start advocating a tracker/ticketing solution for the project which is an interesting idea (note that the sysadmin team already has simple tracker/ticket solution internally). As for providing status updates and stuff like that - this is probably much more than we can do(we don't have unlimited resources ...) - this is still a community project( for 99% of the issues (because those are either fixed very quickly or have no or no noticable effect on the infrastructure). For planned maintenance and reboots/updates of major infrastructure components we are actually trying to inform in advance (but we could certainly improve in that area) > > Right now, when email is down, I just start IM'ing people. A more > structured system would help me. Some people would say a more structured approach to patch/bug tracking would help them (but maybe not help you) ;-) Getting you sent the alerts too is simple to do - but the question is more of what value that information would be for you ? Stefan
On Mon, 24 Mar 2008 19:31:16 +0100 "Andreas 'ads' Scherbaum" <adsmail@wars-nicht.de> wrote: > Yes, i was asked more than once by ppl using a Google talk account. > Would be nice to have them added without adding another account in > pidgin. > Apparently I have more work to do than anyone else :P. I have zero desire to have people contacting me from gmail, yahoo or other such thing. I am busy enough. The purpose of the jabber server was to allow contributors direct access to each other. If we open it up, that has the potential for even more communication overload. If we really want a public jabber server we can do that but I would like it to be separate from the internal jabber.postgresql server. We could then setup the public jabber server so that it can communicate with the private jabber server. This would allow people who want to stay off the public im channels to stay off but still be able to communicate with those that wish to have a public jabber.postgresql.org account. Joshua D. Drake -- The PostgreSQL Company since 1997: http://www.commandprompt.com/ PostgreSQL Community Conference: http://www.postgresqlconference.org/ Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate PostgreSQL SPI Liaison | SPI Director | PostgreSQL political pundit
Stefan Kaltenbrunner wrote: > > > > Right now, when email is down, I just start IM'ing people. A more > > structured system would help me. > > Some people would say a more structured approach to patch/bug tracking > would help them (but maybe not help you) ;-) > Getting you sent the alerts too is simple to do - but the question is > more of what value that information would be for you ? When email was down this past week, I had no idea if the problem was known and who was working on it. I envisioned something like the emergency broadcast system in the USA that sends out an alerts and allows communication to continue while community email is down. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Joshua D. Drake wrote: > This would allow people who want to stay off the public im channels to > stay off but still be able to communicate with those that wish to have > a public jabber.postgresql.org account. I'm not sure I follow this argument. If people don't want to be contacted by outer people, just don't add them as contacts. -- Alvaro Herrera http://www.CommandPrompt.com/ The PostgreSQL Company - Command Prompt, Inc.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Mon, 24 Mar 2008 16:23:19 -0300 Alvaro Herrera <alvherre@commandprompt.com> wrote: > Joshua D. Drake wrote: > > > This would allow people who want to stay off the public im channels > > to stay off but still be able to communicate with those that wish > > to have a public jabber.postgresql.org account. > > I'm not sure I follow this argument. If people don't want to be > contacted by outer people, just don't add them as contacts. > Well I could be wrong on this but my understanding is that not adding them to contacts does not prevent them from trying to be contacted. One of the major reasons people are fleeing yahoo messenger (and why I dropped it years ago) is the amount of people sending me things like: SBF18 36/24/36 I am entirely too old to be futzing with that. Not to mention I am married and it was likely some man sending the message in the first place. Instant messaging is a collaborative work tool for me. I don't want to be contacted with: Hey... your that postgresql dude aren't you? Hey... can I ask you a quick autovacuum question, you work with Alvaro right? Hey... What does Devrim really look like, he sounds cute? Sincerely, Joshua D. Drake - -- The PostgreSQL Company since 1997: http://www.commandprompt.com/ PostgreSQL Community Conference: http://www.postgresqlconference.org/ United States PostgreSQL Association: http://www.postgresql.us/ Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFH6BDHATb/zqfZUUQRAjV/AJ95bnAhqGWkfl7EIeZAKWnOyzf0VwCffdhf SpybMdX2FxXjfEBvNkFTlHk= =qaBh -----END PGP SIGNATURE-----
Hi, On Mon, 2008-03-24 at 13:36 -0700, Joshua D. Drake wrote: > Hey... What does Devrim really look like, he sounds cute? Yes I do. :-P -- Devrim GÜNDÜZ , RHCE PostgreSQL Replication, Consulting, Custom Development, 24x7 support Managed Services, Shared and Dedicated Hosting Co-Authors: plPHP, ODBCng - http://www.commandprompt.com/
Joshua D. Drake wrote: > Well I could be wrong on this but my understanding is that not adding > them to contacts does not prevent them from trying to be contacted. > > One of the major reasons people are fleeing yahoo messenger (and why I > dropped it years ago) is the amount of people sending me things like: > > SBF18 36/24/36 > > I am entirely too old to be futzing with that. Not to mention I > am married and it was likely some man sending the message in the > first place. Instant messaging is a collaborative work tool for me. I > don't want to be contacted with: > > Hey... your that postgresql dude aren't you? > Hey... can I ask you a quick autovacuum question, you work with Alvaro > right? > Hey... What does Devrim really look like, he sounds cute? Funny, I had this exact problem with Yahoo and found the answer last week. In Pidgin under Tools/Privacy, if I choose Yahoo, I can select "Allow only the users on my buddy list" and that should take care of troll IM's. Jabber also has that option. I assume all IM tools have that option. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Joshua D. Drake wrote: > One of the major reasons people are fleeing yahoo messenger (and why I > dropped it years ago) is the amount of people sending me things like: > > SBF18 36/24/36 Hmm, I have no idea what these numbers mean (I can guess), but I think I only got annoyed once via IM and then I learned to set the privacy setting. -- Alvaro Herrera http://www.CommandPrompt.com/ PostgreSQL Replication, Consulting, Custom Development, 24x7 support
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Just got back online ... since its so rare (as often as JD and I agree), just want to make a slight amendment to my original ... I think Bruce's idea has merit, just don't htink having it go through a personal desktop is the answer ... will bring it up on the sysadmins mailing list though ... - --On Monday, March 24, 2008 11:47:20 -0400 Bruce Momjian <bruce@momjian.us> wrote: > Marc G. Fournier wrote: >> > I would like the requirements defined by the -www and sysadmins teams. >> > I don't want us to go off and create yet more infrastructure without >> > full documentation and understanding of the purpose and whether it is >> > redundant. >> > >> > This list is currently redundant as we have the resources and >> > technology already in place to deal with this. I think you are >> > attacking the wrong problem Bruce. >> >> For how often it happens, its worth nothing: I agree 100% with Joshua >> here, but, if Bruce wants to create a list on his personal domain (which, >> in itself is a failure point as it requires that *his* domain is not 100% >> perfect), it doesn't mean that anyone actually has to *use* it ... I know >> I never will, since there already exist faster, more reliable, contact >> methods ... > > I think we need a more centralized contact system than we have now. > Personally when I don't get email I am unsure who to contact (because > community email is down). The email list was a method to centralize > that. The list of possible contact methods listed confirms my point. > > -- > Bruce Momjian <bruce@momjian.us> http://momjian.us > EnterpriseDB http://postgres.enterprisedb.com > > + If your life is a hard drive, Christ can be your backup. + - -- Marc G. Fournier Hub.Org Hosting Solutions S.A. (http://www.hub.org) Email . scrappy@hub.org MSN . scrappy@hub.org Yahoo . yscrappy Skype: hub.org ICQ . 7615664 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQFH6Ckr4QvfyHIvDvMRAhVYAJ4vhlXkYxmiIVw0fvcg/pB9CbFfwQCgjcmU 3AucDB9OAeq82Q7ct1Aui5E= =o9to -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 - --On Monday, March 24, 2008 08:53:26 -0700 "Joshua D. Drake" <jd@commandprompt.com> wrote: > The problem is one of complexity of environment and resources to manage > that environment. This is already being addressed on the sysadmins list > and I expect that it will be resolved within 2 weeks. Does that sound > about right Marc? That is the timeline I proposed, yes ... - -- Marc G. Fournier Hub.Org Hosting Solutions S.A. (http://www.hub.org) Email . scrappy@hub.org MSN . scrappy@hub.org Yahoo . yscrappy Skype: hub.org ICQ . 7615664 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQFH6CuU4QvfyHIvDvMRAoS0AKCsPdMF3TAlYCdFRazs7AtzWgpQjQCbBhKd Re1AvFHiIAoZ6RuYy4wWns8= =nMXL -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 - --On Monday, March 24, 2008 08:57:14 -0700 "Joshua D. Drake" <jd@commandprompt.com> wrote: > Your jabber account can not talk to jdrake@jabber.postgresql.org when > there is a problem with archives. Marc should really be on it too but > besides you too, all of the sysadmins and -www team is already on it. Let me know what I need to connect with ... - -- Marc G. Fournier Hub.Org Hosting Solutions S.A. (http://www.hub.org) Email . scrappy@hub.org MSN . scrappy@hub.org Yahoo . yscrappy Skype: hub.org ICQ . 7615664 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQFH6CvO4QvfyHIvDvMRAuFAAJ4hO1epqu3CW6ylOzCZoSEGu2yZ1wCglcmP /wpHJ3cewEBtNk3M4rbm7k8= =GQt4 -----END PGP SIGNATURE-----
Marc G. Fournier wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > > Just got back online ... since its so rare (as often as JD and I agree), just > want to make a slight amendment to my original ... > > I think Bruce's idea has merit, just don't htink having it go through a > personal desktop is the answer ... will bring it up on the sysadmins mailing > list though ... I would rather see it hosted somewhere else but I don't like to ask someone to set something up if I can set it up myself. If someone else sets it up, that is even better. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 - --On Monday, March 24, 2008 17:52:37 +0000 Dave Page <dpage@pgadmin.org> wrote: > pgsysadmins@agliodbs.com > > And no, I don't think it's ever been used because in the event of > major problems Magnus, Stefan and I communicate via MSN, Jabber, email > or phone - and when Marc is needed and is asleep (which tends to be > the case at the time when things generally seem to go wrong), we page > him. Plus, in my case, if there is a problem, chances are its affecting my personal email too :) Paging, in my case, should always be first method of contact .. MSN/Yahoo second (will get signed up with the jabber.postgresql.org as soon as someone tells me how) ... - -- Marc G. Fournier Hub.Org Hosting Solutions S.A. (http://www.hub.org) Email . scrappy@hub.org MSN . scrappy@hub.org Yahoo . yscrappy Skype: hub.org ICQ . 7615664 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQFH6DAK4QvfyHIvDvMRAlasAKDbeHlUU2WkPEYq7FSjS8E34qeu0gCfbzis JSNLlWbwWOCPSiCL5fS4FFw= =9Mfh -----END PGP SIGNATURE-----
Marc, apparently the signature separators on Majordomo are now broken. I am seeing a single dash and no space: > > - > Sent via pgsql-www mailing list (pgsql-www@postgresql.org) > To make changes to your subscription: > http://www.postgresql.org/mailpref/pgsql-www -- Alvaro Herrera http://www.CommandPrompt.com/ The PostgreSQL Company - Command Prompt, Inc.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Fixed ... and I think I know why (raw vs formatted config in same file) ... when I modified the delivery_rules, it regenerated the 'formatted version' of the config file, which is what I modified last time ... just fixed the 'raw' formatting also, so we should be good through other mods to delivery_rules ... - --On Monday, March 24, 2008 19:53:42 -0300 Alvaro Herrera <alvherre@commandprompt.com> wrote: > > Marc, apparently the signature separators on Majordomo are now broken. > I am seeing a single dash and no space: > >> >> - >> Sent via pgsql-www mailing list (pgsql-www@postgresql.org) >> To make changes to your subscription: >> http://www.postgresql.org/mailpref/pgsql-www > > > -- > Alvaro Herrera http://www.CommandPrompt.com/ > The PostgreSQL Company - Command Prompt, Inc. - -- Marc G. Fournier Hub.Org Hosting Solutions S.A. (http://www.hub.org) Email . scrappy@hub.org MSN . scrappy@hub.org Yahoo . yscrappy Skype: hub.org ICQ . 7615664 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQFH6DN64QvfyHIvDvMRAkcWAJ9ysFcjqAIEkZaCAg+bnrwgmAg91wCeI/px 0e7gyefFA9j8TGsiOiv2cSE= =x8Wv -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 - --On Monday, March 24, 2008 15:04:39 -0400 Bruce Momjian <bruce@momjian.us> wrote: > When email was down this past week, I had no idea if the problem was > known and who was working on it. Trust in the fact that by the time you would notice a problem, we're already working on fixing it ... Stefan has everything monitored automatically, which sends out auto-notices to 'those that can fix it' ... by the time you would notice it, we're already deep into fixing it ... - -- Marc G. Fournier Hub.Org Hosting Solutions S.A. (http://www.hub.org) Email . scrappy@hub.org MSN . scrappy@hub.org Yahoo . yscrappy Skype: hub.org ICQ . 7615664 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQFH6Gin4QvfyHIvDvMRAjwZAJ9CaHzNL+mHu41Bf4uvluJibnUqgwCg6L2c JmEhEnR6Bec3IygG0s/v7UA= =tU+y -----END PGP SIGNATURE-----
Dave, > pgsysadmins@agliodbs.com Heh, I don't even remember setting this up. Better test it; I don't know who's subscribed anymore. -- Josh Berkus PostgreSQL @ Sun San Francisco
On Mon, 24 Mar 2008 23:51:19 -0300 Marc G. Fournier wrote: > - --On Monday, March 24, 2008 15:04:39 -0400 Bruce Momjian <bruce@momjian.us> > wrote: > > > When email was down this past week, I had no idea if the problem was > > known and who was working on it. > > Trust in the fact that by the time you would notice a problem, we're already > working on fixing it ... Stefan has everything monitored automatically, which > sends out auto-notices to 'those that can fix it' ... by the time you would > notice it, we're already deep into fixing it ... So i think the problem is not that you (sysadmin team) already know. The problem is that the others does not know that you know. Try to communicate that you are working on a specific problem. One possible solution: a status page, was already mentioned earlier. Kind regards -- Andreas 'ads' Scherbaum German PostgreSQL User Group European PostgreSQL User Group - Board of Directors
On Tuesday 25 March 2008 13:44, Andreas 'ads' Scherbaum wrote:
> On Mon, 24 Mar 2008 23:51:19 -0300 Marc G. Fournier wrote:
> > - --On Monday, March 24, 2008 15:04:39 -0400 Bruce Momjian
> > <bruce@momjian.us>
> >
> > wrote:
> > > When email was down this past week, I had no idea if the problem was
> > > known and who was working on it.
> >
> > Trust in the fact that by the time you would notice a problem, we're
> > already working on fixing it ... Stefan has everything monitored
> > automatically, which sends out auto-notices to 'those that can fix it'
> > ... by the time you would notice it, we're already deep into fixing it
> > ...
>
> So i think the problem is not that you (sysadmin team) already know.
> The problem is that the others does not know that you know.
>
> Try to communicate that you are working on a specific problem. One
> possible solution: a status page, was already mentioned earlier.
>
>
I have a couple of thoughts wrt this. AUIU we're using nagios, it should be
possible to set up sysadmins and bruce with nagios accounts so we can login
and view the pretty colors that explain what is up or down.
Second, as I think about it, I think the only member of our sysadmin team who
has ops on IRC is Joshua. Perhaps we should see about having a number of us
get ops so we can post notices there. It won't reach everyone, but would hit
a large % of people, and the network is completly seperate in all ways from
the postgresql.org infrastructure.
--
Robert Treat
Build A Brighter LAMP :: Linux Apache {middleware} PostgreSQL
Robert Treat wrote: > > Try to communicate that you are working on a specific problem. One > > possible solution: a status page, was already mentioned earlier. > > > > > > I have a couple of thoughts wrt this. AUIU we're using nagios, it should be > possible to set up sysadmins and bruce with nagios accounts so we can login > and view the pretty colors that explain what is up or down. Yes, Magnus has already set me up with a nagio account. > Second, as I think about it, I think the only member of our sysadmin team who > has ops on IRC is Joshua. Perhaps we should see about having a number of us > get ops so we can post notices there. It won't reach everyone, but would hit > a large % of people, and the network is completly seperate in all ways from > the postgresql.org infrastructure. I have IRC ops too. But I am really looking for something that notifies me that there is a problem, rather than having me go looking for information. If email is down, people should be able to subscribe to alerts and status reports. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 - --On Tuesday, March 25, 2008 21:59:55 -0400 Bruce Momjian <bruce@momjian.us> wrote: > I have IRC ops too. But I am really looking for something that notifies > me that there is a problem, rather than having me go looking for > information. If email is down, people should be able to subscribe to > alerts and status reports. 'k, I can guarantee you that if/when there is a problem, my focus is going to be on *fixing* the problem, not sending out status messages to the effect of "its still down" ... - -- Marc G. Fournier Hub.Org Hosting Solutions S.A. (http://www.hub.org) Email . scrappy@hub.org MSN . scrappy@hub.org Yahoo . yscrappy Skype: hub.org ICQ . 7615664 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQFH6dR54QvfyHIvDvMRAlwdAJ9HKGHe4RdPfDKMzB9Ttjv8S7jqOwCcDU+L CY7bAxcL5OQ43c1CnCOkzKg= =1aN3 -----END PGP SIGNATURE-----
Marc G. Fournier wrote: >> I have IRC ops too. But I am really looking for something that notifies >> me that there is a problem, rather than having me go looking for >> information. If email is down, people should be able to subscribe to >> alerts and status reports. > > 'k, I can guarantee you that if/when there is a problem, my focus is going to > be on *fixing* the problem, not sending out status messages to the effect of > "its still down" ... Any reliable network infrastructure will have the ability for the users of that infrastructure to get a status update on a particular problem. I am able to do this even with my uber cheap DSL line. I call the status line when there is a problem and a nice little attendant says, "Yo... we know there is a problem in hood river, ETA to fix, 2 hours". Do I think it is reasonable that Marc or anyone else start sending out status messages? No. Do I think it is reasonable that when there is a problem that a ticket be created on pmt.postgresql.org that describes the problem, and when the problem is fixed (or if the problem is going to be longer term), that ticket be updated and closed? Yes. We are either a network of professionals supporting an infrastructure, or were not. Whether we like it or not, the moment we agreed to be on sysadmins@, postgresql.org became our customer. We should act like they are our customer and provide the relevant and appropriate customer communications. That means, tickets, updates, communication, and documentation. Sincerely, Joshua D. Drake
Bruce Momjian wrote: > I have IRC ops too. But I am really looking for something that notifies > me that there is a problem, rather than having me go looking for > information. If email is down, people should be able to subscribe to > alerts and status reports. Nagios can be very noisy, however we could have a "status" email list that nagios sent to. I warn you though, it can be very... very noisy. Joshua D. Drake
On Tue, 25 Mar 2008 22:47:03 -0700 Joshua D. Drake wrote: > Bruce Momjian wrote: > > > I have IRC ops too. But I am really looking for something that notifies > > me that there is a problem, rather than having me go looking for > > information. If email is down, people should be able to subscribe to > > alerts and status reports. > > Nagios can be very noisy, however we could have a "status" email list > that nagios sent to. I warn you though, it can be very... very noisy. A filter can be very effective here ;-) At least ppl cannot blame someone that they don't get information about a problem. So if nagios is noisy it's up to the receiver to handle the messages. Kind regards P.S.: and yes, i know, how noisy it can be, the sms inbox on my mobile tells stories - but that's better than not knowing what's going on. -- Andreas 'ads' Scherbaum German PostgreSQL User Group European PostgreSQL User Group - Board of Directors
On Tue, 25 Mar 2008 21:54:22 -0400 Robert Treat wrote: > Second, as I think about it, I think the only member of our sysadmin team who > has ops on IRC is Joshua. Perhaps we should see about having a number of us > get ops so we can post notices there. Sorry for abusing this thread: does someone know from mind how to register the other PostgreSQL channels on IRC if there is no op around? That's not yet a problem, but i don't want to wait if we need an op around. Kind regards -- Andreas 'ads' Scherbaum German PostgreSQL User Group European PostgreSQL User Group - Board of Directors
On Mon, 2008-03-24 at 12:01 -0700, Joshua D. Drake wrote: > On Mon, 24 Mar 2008 19:31:16 +0100 > "Andreas 'ads' Scherbaum" <adsmail@wars-nicht.de> wrote: > > > Yes, i was asked more than once by ppl using a Google talk account. > > Would be nice to have them added without adding another account in > > pidgin. > > > > Apparently I have more work to do than anyone else :P. I have zero > desire to have people contacting me from gmail, yahoo or other such > thing. I am busy enough. > > The purpose of the jabber server was to allow contributors direct > access to each other. If we open it up, that has the potential for even > more communication overload. > > If we really want a public jabber server we can do that but I would > like it to be separate from the internal jabber.postgresql server. We > could then setup the public jabber server so that it can communicate > with the private jabber server. If people want the public chat, why don't they just use gtalk or whatever public service they prefer? I don't see a reason for us to run one private and one public server. In fact, I don't see a reason for us to run a public one at all - if we were running that, we could just as well run nothing. AFAIK every reasonable IM client people would use today can easily add two jabber accounts (or more), so I don't see the problem, really. //Magnus
On Wed, Mar 26, 2008 at 11:39:33AM +0100, Magnus Hagander wrote: > I don't see a reason for us to run one private and one public server. In I agree with that. The problem is that the protocol is actually designed so that you don't need these infernal private servers. > AFAIK every reasonable IM client people would use today can easily add > two jabber accounts (or more), so I don't see the problem, really. Part of the point of the Jabber protocol is to avoid having to track which account you have to use to talk to which people. Having all these private servers is taking us back to the days of GEnie, CompuServe, and other such closed networks: you had to know which gateway (or series of gateways, in the case of UUCP mail) to use in order to reach your party. Computers are better at routing than humans. A