Обсуждение: Email monitoring
Should we set up a dummy email list that I can send to in an automated way every hour? If I don't get a reply in X minutes I can send out an alert to people. -- 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 11:17:58AM -0400, Bruce Momjian wrote: > Should we set up a dummy email list that I can send to in an automated > way every hour? If I don't get a reply in X minutes I can send out an > alert to people. What is the problem you are trying to solve? A
Andrew Sullivan wrote: > On Mon, Mar 24, 2008 at 11:17:58AM -0400, Bruce Momjian wrote: > > Should we set up a dummy email list that I can send to in an automated > > way every hour? If I don't get a reply in X minutes I can send out an > > alert to people. > > What is the problem you are trying to solve? Email is blocked but there is no notification except "Hey, I haven't gotten email in a while". -- 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 11:26:03AM -0400, Bruce Momjian wrote: > > What is the problem you are trying to solve? > > Email is blocked but there is no notification except "Hey, I haven't > gotten email in a while". Adding a completely different address for that is not really a good monitor. There are at least these problems possible: - routing problems - mail server problems - list server problems - machine failure problems - delivery problems These each need different monitors, I suggest. Let's work on those. A
Andrew Sullivan wrote: > On Mon, Mar 24, 2008 at 11:26:03AM -0400, Bruce Momjian wrote: > > > What is the problem you are trying to solve? > > > > Email is blocked but there is no notification except "Hey, I haven't > > gotten email in a while". > > Adding a completely different address for that is not really a good monitor. > > There are at least these problems possible: > > - routing problems > - mail server problems > - list server problems > - machine failure problems > - delivery problems > > These each need different monitors, I suggest. Let's work on those. OK, but I have no idea how to monitor those, just email send/receive. -- 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, Bruce Momjian wrote: > Andrew Sullivan wrote: >> On Mon, Mar 24, 2008 at 11:17:58AM -0400, Bruce Momjian wrote: >>> Should we set up a dummy email list that I can send to in an automated >>> way every hour? If I don't get a reply in X minutes I can send out an >>> alert to people. >> >> What is the problem you are trying to solve? > > Email is blocked but there is no notification except "Hey, I haven't > gotten email in a while". 'k, well, all this would do is test if mail was blocked to you, but not necessarily to everyone else ...
Marc G. Fournier wrote: > > > On Mon, 24 Mar 2008, Bruce Momjian wrote: > > > Andrew Sullivan wrote: > >> On Mon, Mar 24, 2008 at 11:17:58AM -0400, Bruce Momjian wrote: > >>> Should we set up a dummy email list that I can send to in an automated > >>> way every hour? If I don't get a reply in X minutes I can send out an > >>> alert to people. > >> > >> What is the problem you are trying to solve? > > > > Email is blocked but there is no notification except "Hey, I haven't > > gotten email in a while". > > 'k, well, all this would do is test if mail was blocked to you, but not > necessarily to everyone else ... True, but I thought that would help. Historically, I thought most failures were for all users. -- 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. +
Hello, On Mon, 24 Mar 2008 11:17:58 -0400 (EDT) Bruce Momjian wrote: > Should we set up a dummy email list that I can send to in an automated > way every hour? If I don't get a reply in X minutes I can send out an > alert to people. That could be (if needed) easier done with the existing monitoring system. But i'm all for a working mailinglist solution instead some workarounds to check a non-functional setup. Bruce: if you setup such a system you (but nobody else) knows about a problem. What use will be this system, if you are offline, away or just sleeping? So we need some more people sending mails periodically and check the result. This will result in a lot overhead and i assume in some false (and some true) positives. Kind regards -- Andreas 'ads' Scherbaum German PostgreSQL User Group European PostgreSQL User Group - Board of Directors
Andreas 'ads' Scherbaum wrote: > > Hello, > > On Mon, 24 Mar 2008 11:17:58 -0400 (EDT) Bruce Momjian wrote: > > > Should we set up a dummy email list that I can send to in an automated > > way every hour? If I don't get a reply in X minutes I can send out an > > alert to people. > > That could be (if needed) easier done with the existing monitoring > system. But i'm all for a working mailinglist solution instead some > workarounds to check a non-functional setup. > > Bruce: if you setup such a system you (but nobody else) knows about a > problem. What use will be this system, if you are offline, away or just > sleeping? So we need some more people sending mails periodically and > check the result. This will result in a lot overhead and i assume in > some false (and some true) positives. I assumed a failure would be reported to the new pgsql-emergency email 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. +
On Mon, 24 Mar 2008 11:44:50 -0400 (EDT) Bruce Momjian wrote: > I assumed a failure would be reported to the new pgsql-emergency email > list. Who will do the report? Bye -- Andreas 'ads' Scherbaum German PostgreSQL User Group European PostgreSQL User Group - Board of Directors
Andreas 'ads' Scherbaum wrote: > On Mon, 24 Mar 2008 11:44:50 -0400 (EDT) Bruce Momjian wrote: > > > I assumed a failure would be reported to the new pgsql-emergency email > > list. > > Who will do the report? A script would send out the email on failure. -- 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: > Andreas 'ads' Scherbaum wrote: >> On Mon, 24 Mar 2008 11:44:50 -0400 (EDT) Bruce Momjian wrote: >> >>> I assumed a failure would be reported to the new pgsql-emergency email >>> list. >> Who will do the report? > > A script would send out the email on failure. we already have a quite advanced and capable monitoring solution that can (and is) easily do this and much more - I totally fail to see why we need to add another ad-hoc solution that will not be able to detect a large number of problems ... Stefan
Andrew Sullivan wrote: > On Mon, Mar 24, 2008 at 11:26:03AM -0400, Bruce Momjian wrote: >>> What is the problem you are trying to solve? >> Email is blocked but there is no notification except "Hey, I haven't >> gotten email in a while". > > Adding a completely different address for that is not really a good monitor. > > There are at least these problems possible: > > - routing problems > - mail server problems > - list server problems > - machine failure problems > - delivery problems > > These each need different monitors, I suggest. Let's work on those. we actually do have monitoring for most if not all of the above mentioned things and more ... Stefan
On Mon, Mar 24, 2008 at 3:36 PM, Andrew Sullivan <ajs@crankycanuck.ca> wrote: > On Mon, Mar 24, 2008 at 11:26:03AM -0400, Bruce Momjian wrote: > > > What is the problem you are trying to solve? > > > > Email is blocked but there is no notification except "Hey, I haven't > > gotten email in a while". > > Adding a completely different address for that is not really a good monitor. > > There are at least these problems possible: > > - routing problems > - mail server problems > - list server problems > - machine failure problems > - delivery problems > We already have perfectly good monitors for most of that, sending alerts outside of the postgresql.org mail infrastructure. -- Dave Page EnterpriseDB UK Ltd: http://www.enterprisedb.com PostgreSQL UK 2008 Conference: http://www.postgresql.org.uk
Dave Page wrote: > On Mon, Mar 24, 2008 at 3:36 PM, Andrew Sullivan <ajs@crankycanuck.ca> wrote: > > On Mon, Mar 24, 2008 at 11:26:03AM -0400, Bruce Momjian wrote: > > > > What is the problem you are trying to solve? > > > > > > Email is blocked but there is no notification except "Hey, I haven't > > > gotten email in a while". > > > > Adding a completely different address for that is not really a good monitor. > > > > There are at least these problems possible: > > > > - routing problems > > - mail server problems > > - list server problems > > - machine failure problems > > - delivery problems > > > > We already have perfectly good monitors for most of that, sending > alerts outside of the postgresql.org mail infrastructure. Oh, OK. -- 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 04:51:38PM +0000, Dave Page wrote: > > We already have perfectly good monitors for most of that, sending > alerts outside of the postgresql.org mail infrastructure. This is what I thought. I want to avoid adding monitors for things that are already monitored at a lower level, because it just means more alerts with no additional benefit. A