The Easy Poll

  • YesYesYesYesYES... On both.

    How many times have we gone out to the websites looking for the solution to a problem and seen the line "This fix is only for X problem, don't install if your mother has blue hair and the bathroom sink is backing up because the fix may cause more problems"? (Or something of the sort).

    They don't put that warning on service packs and the fact that they put it on hot fixes tells me that they don't trust the hot fix code to completely resolve the issue without causing more problems. So then why why why are they not verifying this code and rolling it up in a Service Pack?

    Then there's the problem of the server admin who have bad hot fix experiences so refuse absolutely to put in a hot fix unless it's either critical or a service pack. And in a big corporation, SQL Server ends up getting the short end of the patch fixing. Oh, they'll patch the OS that SS is installed on, but we'll wait until the patches are all put in an SP before we fix SQL Server. I can't blame them. I don't want to install a patch, bring the server down and spend the next 72 hours rebuilding it back to it's pre-patch status.

    So why isn't Microsoft aware of this behavior in the marketplace? Yes they can make money selling to the non-tech people on the basis that the software will fix all the problems (There's a Dogbert storyline in this), but if the techs refuse to apply the software or keep complaining, MS is eventually going to lose money. Look at all the people fleeing to Open Source.

    So, yes, absolutely. They should provide regular service packs and always give a "final" service pack for which ever product they're about to stop supporting. With 2008 being pushed further and further back, they're going to have to do something about 2000 or risk a lot of marketplace anger for dropping 2000 support and not truly supporting 2005.

    Thanks my $.02 rant. Thanks for letting me borrow your soapbox.

    Brandie Tarvin, MCITP Database AdministratorLiveJournal Blog: http://brandietarvin.livejournal.com/[/url]On LinkedIn!, Google+, and Twitter.Freelance Writer: ShadowrunLatchkeys: Nevermore, Latchkeys: The Bootleg War, and Latchkeys: Roscoes in the Night are now available on Nook and Kindle.

  • Brandie Tarvin (2/1/2008)


    They don't put that warning on service packs and the fact that they put it on hot fixes tells me that they don't trust the hot fix code to completely resolve the issue without causing more problems. So then why why why are they not verifying this code and rolling it up in a Service Pack?

    Actually - I thought that was exactly what they do. They put the hot fix out with the warning WHILE they're incorporating it into the next service pack. Doesn't help the timing problem through...

    I hope that doesn't kick the old soapbox out from under ya...I'll be needing that later today.

    ----------------------------------------------------------------------------------
    Your lack of planning does not constitute an emergency on my part...unless you're my manager...or a director and above...or a really loud-spoken end-user..All right - what was my emergency again?

  • Yes, we need SP3 for SQL 2005! I had to jump through hoops to get CU4 for SP2 from Microsoft, because my security token cache keeps growing like crazy. This to me is a serious issue as it lead to our production server blue screening, and my only choice was to patch it with the cumulative update with the disclaimer that they are not responsible if it makes your situation worse. The fact that they do not release these important fixes in a more frequent and reliable fashion, makes me lose faith in the product. It also took me 5 separate phone calls to get the patch, which really got me fired up. The funny thing is a quality assurance agent from Microsoft called later to survey me about my experience (so I could get on my soapbox) and five minutes into the survey...."I'm sorry sir, I will have to call you back, my computer just crashed and won't come back up!"

  • I absolutely agree that MS should release SP3 for 2005. Either that, or the should put a little more effort into testing an polishing up the CU packages so they can replace SPs. Who wouldn't like 6 SPs a year?

    Here's a question I posed to a MS SQL Server support representative on 1/29/08:

    Do you know if SP3 is on the near horizon for

    Sql Server 2005? Or are we going to see several more cumulative patches

    before then. Seems like it's been about a year since SP2, so I thought

    I'd ask.

    And the response received:

    At this point with the success of the Cumulative Updates for SQL Server 2005 Service Pack 2 and the push to get SQL Server 2008 released I am not sure if or when Service Pack 3 would be released.

    We have been working with MS for over a month to get an very serious issue with snapshot isolation in 2005 fixed so that udfs don't lock up a process when called. That's right, you can't event call a multi-statement udf in snapshot isolation without risking a deadlocked spid that cannot be killed without restarting the server.

    Now, I must give MS credit in that they have created a hotfix for us (and the fix should/may appear in CU6, which should release in mid Feb). However, without full regression testing, and due to the fact that installing the update in our test environment didn't go so hot, we're having a hard time getting this hotfix approved for install on out production servers.

    According to Microsoft:

    Yes this fix has been regression tested but only for functionality that is expected to be affected by your fix. Full regression testing is only performed on major releases and Service Packs.

    I am not aware of any recent issues about installing a Cumulative Update on a server that has a fix installed, but it is still possible. Yes we do test this but the only true test is for you to try it in a test environment.

    Due to the lack of full regression testing for cumulative updates and hotfixes, it's very difficult for us to justify installing such an update on our production servers. We do have test environments, yes, but we don't have the resources to devote to fully retesting a product that should be _very_ stable by now every time a new update is released. We just can't afford to suffer any downtime due to lack of testing at MS!

  • YES, I want SQL Server 2000 service pack 5, 6, 7, ...

  • I like what the guy on the Dyson adverts says: "I just think that things should work properly." I agree with majorbloodnock. I don't much care how that gets done.

    As for the state of things well I harp on the Windows Task Scheduler. Look a the number of replacements for it. I even wrote one here at DCT. It's broken, Redmond knows it, and they have said that they have no intention fixing it. Now watch the flood of posts that will follow from folks saying they have no problem with WTS. Note the ones from people who find WTS so unstable as to be useless.

    See a parallel to MS-SQL?

    ATBCharles Kincaid

  • I agree with you whole heartedly, with explicitives. Yes to 2000 SP5. Yes to 2005 SP3. Almost sound like ballot measures, don't they?! :w00t:

    I'm more concerned about getting all the patches I can for 2005, as this is the year we take our antiquated ERP system (which can't upgrade past 2000 SP3), and upgrade it onto 2005. Having all the patches rolled into a cumulative update gives everyone a well defined starting point.

  • Matt Miller said:

    They put the hot fix out with the warning WHILE they're incorporating it into the next service pack. Doesn't help the timing problem through...

    Use SQLServerCentral;

    GO

    Set Rant_Warning ThisThread On;

    I know, but in this case it seems they're going leave the warning on it (because they don't want to do more work on the problem) and never roll this stuff up into a Service Pack. And as multiple people have said right after you and I posted, no techie wants to be responsible for bringing the production systems down on an untested product.

    If there were a timing issue, i.e. Microsoft actually was planning on doing an SP3 but it was going to be late, that's one thing. This, unfortunately, seems to be a case of "We'll be making too much money on the new SQL Server version to care about your silly little problems".

    Set Rant_Warning ThisThread Off;

    I hope that doesn't kick the old soapbox out from under ya...I'll be needing that later today

    I don't mind. It's Steve's soapbox. Let him chase after it. @=)

    Besides, I get workman's comp at my workplace. Perfect for hangnails and injuries due to falling from soapboxes. @=)

    Brandie Tarvin, MCITP Database AdministratorLiveJournal Blog: http://brandietarvin.livejournal.com/[/url]On LinkedIn!, Google+, and Twitter.Freelance Writer: ShadowrunLatchkeys: Nevermore, Latchkeys: The Bootleg War, and Latchkeys: Roscoes in the Night are now available on Nook and Kindle.

  • Yes, loose ends should be tied up before moving on.

    SQL 2000 is still being widely used, and let's not mention SQL 2005 SP2. Oh, I just did.

    ML

    ---
    Matija Lah, SQL Server MVP
    http://milambda.blogspot.com

  • I agree completely with Steve ... SP's are expected, we shouldn't need to "ask" for them.

    We're currently on SQL 2005, but back when we had SQL 2000 we tried SP4 but rolled back to SP3 due to the AWE issues. I had always hoped for SP5 to fix the problem, but it never arrived.

    One thing the peeves me about Microsoft is that they'd rather spend $44.6B on yahoo!, then spend $44.6K on providing us with SP's. Perhaps $44.6K is a little less then it really cost, but I think you get the idea. The point is, they tend to release a product and instead of spending the time and money to really make that product rock solid, they move on to the next installment and leave people hanging.

    For full discloser regarding my Microsoft ranting ... I'm a former Microsoft employee and still hold shares of MSFT in an IRA ... but this doesn't mean I need to like what I'm seeing.

  • I have posted this for Microsoft in the MVP area. Hopefully they'll see it.

  • Yes to 2005 SP3. And I want it MUCH more thoroughly tested than SP2 was!!!

    Best,
    Kevin G. Boles
    SQL Server Consultant
    SQL MVP 2007-2012
    TheSQLGuru on googles mail service

  • Definitely YES to both. We're waiting to move our remaining 2000 dbs until 2008 is released, and with testing, etc, it will be the end of the year at the earliest. So we still need to support them for a while. And our 2005 dbs will probably wait until all 2000 dbs are moved before we look to move them to 2008.

  • Unfortunately (for you) or fortunately (for me) no problems encountered with current service pack.....leave it alone...

    Thanks:cool:

  • Add my vote for a SQL 2005 SP3. I do not feel comfortable installing cumulative patches. I have been asking our TAM for 2 months when it will be available, so the requests from custormers are being made.

    We are migrating from SQL2000, so I would not care about a SQL 2000 SP5.

Viewing 15 posts - 31 through 45 (of 72 total)

You must be logged in to reply to this topic. Login to reply