Nightmare !!! (Mac OS X)

[bhwong]

I think it is irresponsible for Zimbra to release a version for an platform without fully tested it to ensure that it works. I will think twice before I upgrade my Zimbra, just to make sure that the pioneers root out all the bugs first.

[phoenix]

Originally Posted by bhwong

I think it is irresponsible for Zimbra to release a version for an platform without fully tested it to ensure that it works.

That's a rather silly statement, of course the product was tested and passed all the QA tests. How do you reckon an intermittent problem in the o/s is dealt with? By definition you can't tell when it's going to fail - once again, this isn't a Zimbra bug.

Originally Posted by bhwong

I will think twice before I upgrade my Zimbra, just to make sure that the pioneers root out all the bugs first.

I hope you'll be going to the Apple forums and chastising them for releasing an operating system with this kind of bug in it. Don't take it too seriously, all software has bugs in it.

[gvillaran]

can u go back to your previous version? so it can work good again?

[kygeek]

Originally Posted by phoenix

That's a rather silly statement, of course the product was tested and passed all the QA tests. How do you reckon an intermittent problem in the o/s is dealt with? By definition you can't tell when it's going to fail - once again, this isn't a Zimbra bug.

I hope you'll be going to the Apple forums and chastising them for releasing an operating system with this kind of bug in it. Don't take it too seriously, all software has bugs in it.

wow that is a silly statement too. Dont take it too seriously ?!? when you have 100 people screaming their email does not work its hard to not be serious about it. And after investigating the issue with DNS on 10.5 i have found that classifying this issue as a bug is sketchy at best. I found numerous results that developers mention having a DNS issue because leopard requests SRV records instead of A records now. but guess what ! those developers had to use work arounds in order to make THEIR software work properly. the bottom line is that if your DNS server or upstream server does not handle SRV records properly then it will cause the os to hang waiting on a request. this could be classified as a bug on the Server end or on the OS end.... depends on how u look at it really. My point is that i needed a fix of somekind. I did not receive that from the support team. A piece of software was put into circulation that has an alpha quality bug in it and the customer suffers. The reason the bug is 'intermittent' is because not all DNS servers are created equal thus making the problem show up sometimes and sometimes not. And it didnt take me a year to research the issue to figure out what MIGHT be the problem.

[kygeek]

i thought about using time machine to restore back to last thursday, but we receive so much volume management decided that would be a nightmare too. I have kept the server availability to 60% by manually starting and stopping it over and over all day long including weekends. I may have no choice but to restore it in its usable state before this mess.

[quanah]

kygeek: I understand your frustration. I've spent numerous hours looking at various Leopard boxes that have run into this problem. The problem is quite clearly with DNS resolution in OSX 10.5 being broken after sudden power outages or reboots due to hard lockups. dscacheutil (an Apple utility, not anything shipped by Zimbra) clearly shows that it is failing to cache entries that should be being cached. I've continued to update the bug I filed with Apple with information as it came available. Until Apple fixes the problem with the OS, there literally is nothing Zimbra can do about it. And I have seen, in my debugging, other non-ZCS applications running on the boxes having problems as well.

I honestly don't know what more Zimbra can do than what we've already been doing, which is to gather information on the problem as it comes available and pass it on to Apple.

--Quanah

[kygeek]

what you can do about it is warn people that if their machine locks up (which by the way mine did not... nor did it lose power... was simply an upgrade... I didnt even reboot the machine) it makes their mail server software useless.
that should be stated on the download page in bold red letters... or just say that OSX is NOT supported until its fixed. then you wouldnt have folks investing in infrastructure needlessly.

[kygeek]

Also i found this

"

The DNS resolver in Leopard has been changed to first attempt SRV requests for lookups initiated by the getaddrinfo() function. If the user’s DNS server drops these requests the DNS lookup may take an extended period of time to complete (30 seconds to several minutes) as Leopard tries different domain requests and eventually falls back to making an A record request. This can result in application freezes or timeouts"

and

It’s important to note that this problem does not appear to be a bug in Leopard – it’s caused by old, buggy, or misconfigured DNS servers. The change in Leopard to use the latest IETF recommendations for DNS lookups is simply bringing the DNS server problem to the surface. It’s unclear how many users or applications will be affected by this change, since it only appears with some DNS servers and only for applications using getaddrinfo (most applications still use gethostbyname). For many users these days, their DNS server is actually their home router, which then proxies the request (possibly with a local cache) – so updating router firmware may address the issue.

[quanah]

I've never seen it occur on an OSX 10.5 system that had not either encountered a hard reboot of some sort. We've extensively tested our upgrade procedures without seeing this ever occur.

Also, if the DNS bits you refer to after this were the issue, we'd always see it for servers, not have it only show up after hard locks.

If you still have your non-functional system up and running, and wanted to give me access, I'd be curious to look at it, as it would be more useful data. You can send me a PM if interested.

[kygeek]

i promise you there was not a hard lock... the only hard lock that happened on this system was AFTER the upgrade and it was yesterday when the zimbra technician was working on the system.