Server reboot

[TedCavitt]

The times on the weblog need to be fixed. I don't know what time zone it's running on but it's not PDT for sure.

This sucks. I don't even check my mail that often. With the amount I'm paying I expect better uptime.

[CML209]

the other thread went into technobabble, and my feeling is that this is the very stuff FM needs to read. During most outages, the FM guys get pats on the backs when it's over and cheering sections along the way. This should not be the case this time. Sorry, but this is more of a pain than I am willing to deal with, yet I must. If I must, so must they.
CML

[TheLastShack]

Yes.... This is bad. I may now just have to settle for a combination of my school's 25mb IMAP box and Yahoo! Mail. If Yahoo! Mail or GMail provide IMAP in the future, either one will be my main choice. I also my decided to set up a website with DreamHost again. They provide great IMAP e-mail, and the SquirrelMail interface.

[bitequator]

Originally posted by severach
If you only have a couple aliases on your FM domain then Yahoo's 5 email accounts are no problem. I'm already working on 30 and plan on more, so that's 25 aliases for which there can be no corresponding Yahoo accounts. Remember, FM creates aliases in your domain. Yahoo creates accounts like a hosting or business service would. Yahoo isn't too keen on giving out 50 100MB accounts just because we ask nicely and there appears to be no option to buy.

Hmm, I'm not sure I'm understanding this correctly. I've never tried the Yahoo Personal Address (own-domain) service, but I thought it gives you 5 aliases on that domain (not 5 mailboxes) -- which I assumed would either be automatically attached to your Yahoo mailbox account (free or paid), or perhaps you can even set each alias to forward to an independent destination (even an external account)?

Originally posted by severach
No, I don't think it was a judgement call. Jeremy already said his backup SMTP was working so if you didn't do that then he would be getting them on a server that you can't browse, not your other service. It's not like the sender sprays the message at all the backup servers when the primary fails. It may have been a judgement call at the time but now we can see that it is the only way that will work. If the first SMTP server is the only one you can browse, than you want mail elsewhere if it can't get to that one.

Actually what I said was I have the 3rd-party MX inserted after "smtp.us2" but before "smtp.eu". If I had inserted it before "smtp.us2", then the result in this outage would've still been the same I think (ie. new emails getting routed to the backup account automatically)...?

I had my reasoning assumptions for that but I forget what they are... I think because both "smtp.us" (the primary) and "smtp.us2" are in the same NYI data center, I figure any problem that would cause the first to go down without affecting the second, would likely only be very minor short-lived outages, so it's not worth having the 3rd-party MX server step in on those very temporary fluctuations. But if both the NYI servers go down (or the whole NYI facility went offline), it could possibly be due to a larger problem leading to a longer downtime, in which case I think it's good to have the 3rd-party MX step in before the "smtp.eu" Texas server so I can receive and read new emails in the backup account immediately...


Edit: I can't recall now, is "smtp.us2" also a store-and-forward (backup queueing) server that forwards to "smtp.us", or can it deliver to the mailstore directly?? Does anyone know?

[Jeremy Howard]

Thanks for pointing out the PDT problem - fixed now. The times are actually UTC.

[TedCavitt]

Fastmail is supposed to be email for power users. If this happens again I'm moving to Runbox.

[CML209]

At least Turner Classic Movies has Bogart films on all night. Twenty minutes to go, right?
cml

[FMRocks]

Ok now the login page loads but can't log in - takes right back to the outage message.

[FMRocks]

Oops... spoke too soon.... webmail is back up now... so is IMAP.

[FMRocks]

Just a heads up... FM is back up now.

[rakhesh]

Originally posted by bitequator
I had my reasoning assumptions for that but I forget what they are... I think because both "smtp.us" (the primary) and "smtp.us2" are in the same NYI data center, I figure any problem that would cause the first to go down without affecting the second, would likely only be very minor short-lived outages, so it's not worth having the 3rd-party MX server step in on those very temporary fluctuations. But if both the NYI servers go down (or the whole NYI facility went offline), it could possibly be due to a larger problem leading to a longer downtime, in which case I think it's good to have the 3rd-party MX step in before the "smtp.eu" Texas server so I can receive and read new emails in the backup account immediately...

Edit: I can't recall now, is "smtp.us2" also a store-and-forward (backup queueing) server that forwards to "smtp.us", or can it deliver to the mailstore directly?? Does anyone know?

Good assumptions.

But something else struck me in the course of this discussion -- if the database is what's down, then why is the "smtp.us" server too down (for mail to go to ur backup servers)? Ofcourse, I assume the database would be separate from the smtp servers -- to keep the load etc separate -- and so if they are both on the same machine, then yeah, I understand why "smtp.us" is down. But then again, even if the database and "smtp.us" are on the same machine (and so "smtp.us" is down), wouldn't "smtp.us2" be up and thus mail not getting delivered to your backup mail servers?

[CML209]

Yes, but mail that was sent to me hours ago, even before the official downtime is still not there. I forwarded something without knowing there was downtime and it's not there.
Of little use yet.
CML
and a mail that was sent since it's been back up is not there yet.

[Jeremy Howard]

Originally posted by rakhesh
But something else struck me in the course of this discussion -- if the database is what's down, then why is the "smtp.us" server too down (for mail to go to ur backup servers)?

The database contains information about what users we have, what server they are on, what aliases they have, etc. It is impossible to deliver mail without the database server, unfortunately.

[rakhesh]

Originally posted by Jeremy Howard
The database contains information about what users we have, what server they are on, what aliases they have, etc. It is impossible to deliver mail without the database server, unfortunately.

Hmm, I see. So when the database is down, the smtp.us server doesn't know where to deliver the mail to etc, and so it doesn't accept mail ... But the smtp.eu server doesn't need to know all these details -- it simply accepts whatever mails are given to it (its a store and forward) -- coz all the actual mail filing is being done by smtp.us anyways. Right?

Where does smtp.us2 fit into this picture? If the database is down, does that too not accept mails? (I know the answer is yes, coz if it had been accepting mails, then they would have gone to bitequator's backup servers ... still, thought I'd confirm).

[oysterquartz]

Originally posted by Jeremy Howard
The database contains information about what users we have, what server they are on, what aliases they have, etc. It is impossible to deliver mail without the database server, unfortunately.

You didn't address the question as to why File Space should be compromised, and ones virtual domain rendered useless for email and web/file access.

IMO this is an inherent flaw in architecture.