locked
KB4015553 and KB4019215 RRS feed

  • Question

  • Hi everyone, a couple weeks ago (after April 21st) we have experienced issues when the update https://support.microsoft.com/en-us/help/4015553/windows-8-1-windows-server-2012-r2-update-kb4015553 was installed on Domain Controlers (Windows SBS 2008, 2008 R2 and 2012 R2)

    symptoms:

    1. ADDS service nos answering properly

    2. DNS issues affects item 1 (ADDS) and so on and so far 

    3. workstations couldn't access network resource due to dns issues (dfs, unc path, etc.)

    when you restart the server sometimes it starts to answer back on network but 99% it fails

    we promptly removed the KB4015553 once we noticed that it came through WSUS but as a Preview of Monthly Rollup

    Now we figured out the MSFT released the May 9, 2017—KB4019215 (Monthly Rollup)

    https://support.microsoft.com/en-us/help/4019215/windows-8-update-kb4019215

    and we're experiencing the same issues

    does anyone has the same symptoms or issues?

    thanks,

     


    Thiago Beier Se foi útil marca como resposta! Don´t forget to mark as answer!

    Monday, May 15, 2017 3:06 PM

All replies

  • Hi Thiago Beier,

    1. After installing KB4015553 or KB4019215, please run sfc /scannow to check if there are corrupted or missing system files.

    2. Since it is a DC, we may also run dcdiag to test if it will report some useful information;

    3. Security-only rollup may include security updates, if monthly rollup has issues, we may test if installing the security-only rollup(KB4019213) could avoid the issue.

    4. What's more, after searching around, seems you are not the only one reporting issues after installing KB4015532, we'll still keep eyes on this issue, if there are any news, we'll update this thread as soon as possible.

    Best Regards,

    Anne


    Please remember to mark the replies as answers if they help.
    If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com.

    Friday, May 19, 2017 2:26 AM
  • https://social.technet.microsoft.com/Forums/windowsserver/en-US/891c3a09-930c-4799-b30f-10f2a4451b5c/performance-issue-after-installing-kb4015553-kb4019215-on-to-server-2012-r2-with-hyperv-role?forum=winserverhyperv

    • Hi

           Having installed KB4015333 last month and having performance issues with our Windows Server 2012 R2 Hyper-V machine whereby within 24 hours the system (ntoskrnl.exe) process would climb to > 16,000 threads and the server would become almost unresponsive. We removed KB4015333 and the problem then went away. This update was superseded and replaced by KB4019215. We have installed this update within the last 24 hours and we are currently observing the same type of behavior. I have restarted our Hyper-V server this morning and with 30 minutes the system process has climbed to 900 threads.

           We are also experiencing the same issue with a Windows Server 2016 standard that we use a file server after installing either the April or May CU. We do have a support case that we have opened with Microsoft for the 2016 problem and I suspect we will need to do the same with the 2012 R2 issue.

           I am keen to know if anyone else is seeing these problems?



    • Edited by Mike ERS Friday, May 19, 2017 7:10 AM
    Friday, May 19, 2017 6:52 AM
  • Having noted Annes response I can confirm the following:

    We installed KB4018271, KB4014590, KB4014556 & KB4014510 on the 16th May and had no issues with the server

    We then installed KB4019213 on the 17th May and again had no issues with the server.

    The only outstanding update for May 2017 for Server 2012 R2 was then KB4019215 this was installed 19th May and now the server has the issue described in my previous post.

    Friday, May 19, 2017 7:19 AM
  • Hi Anne He, we're gonna build an work plan to follow next time we deal with that or to try at a controlled environment and I'll get back here with the results of sfc /scannow and dcdiag 

    regards,

    Thiago


    Thiago Beier Se foi útil marca como resposta! Don´t forget to mark as answer!

    Friday, May 19, 2017 8:55 PM
  • Hi,

    same issue here. 3 different 2012R2 Server (2 x DC, 1 x Fileserver in two different customer sites; all vmware vm).

    After installing updates all Servers partially lose network connection (approx every 20 hours). Shares are not reachable, Internet inside Browsersoftware not reachable. DNS in cmd is ok. Ping the server is ok. Ping from Server not. RDP to Server not possible. Sfc /scannow no negative results.

    In one machine i saw hundreds off processes regarding our remote access app. Server restart is not possible. Only way to temporaly fix it: Hard shut down and power on. 

    2 Servers were fully updated on 11.05.17.....the third one yesterday. The problems started in each case one day later.

    I deinstalled KB4019215. No change in behaviour.

    Common EventID on the DCs before connection lost is:

    1206, Source: ADWS.

    Best Regards

    Ulf


    • Edited by ureitz Sunday, May 21, 2017 9:25 PM
    Sunday, May 21, 2017 9:19 PM
  • Hi,

    We had the same problem. I just solved this, removing the update KB4015553. 
    My server is Windows Server 2012 Datacenter and affected my Failover Cluster. After a period (24h~36h) one of the nodes down.
    I have identified one more update in the same period of this, the KB4014551, I have not yet removed it. It seems that my environment is still unstable.
    For security, I kept pendent the update KB4019215.Anyone knows if the Microsoft pronounced some hotfix?

    Tuesday, May 23, 2017 5:43 PM
  • We first experienced a problem with patches KB4019215 back in April.  Since then there have been a number of posts by a variety of people on Reddit\sysadmin, Microsoft Technet, and Spiceworks.

    Check out this Reddit  thread - it has some of the other thread links in it.  https://www.reddit.com/r/sysadmin/comments/6cum10/headsup_2012_r2_servers_with_iscsi/

    No official word from Microsoft on what happened yet that I have seen.

    Tuesday, May 23, 2017 8:49 PM
  • Hi Anne He, we're gonna build an work plan to follow next time we deal with that or to try at a controlled environment and I'll get back here with the results of sfc /scannow and dcdiag 

    regards,

    Thiago


    Thiago Beier Se foi útil marca como resposta! Don´t forget to mark as answer!

    Hi Thiago Beier,

    Just to check if you got any progress with the issue? Welcome to feedback.

    Best Regards,

    Anne


    Please remember to mark the replies as answers if they help.
    If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com.

    Wednesday, May 24, 2017 2:12 AM
  • We supports numerous companies. We have a physical SQL server running on Windows server standard 2012 experiencing issues as mentioned above. I have looked for the updates mentioned and see kb4019217 which includes kb4019215. Threads for system sitting at 16,549 as I type. Reboot fixes it temporarily.

    I am going to remove this to see if it helps.

    We support another server (Virtual) acting solely as a DC and this is losing network connectivity. We can't connect externally at all and have to force the machine off in HyperV. It doesn't shut down. This also has this update.

    Thanks for giving me those kb numbers. Fingers crossed it works after. I replied mainly to give the new kb number.

    Wednesday, May 24, 2017 9:14 AM
  • We found a bad connection iSCSI in my server.
    We have adjusted this and will wait if the symptom will occur again.
    Wednesday, May 24, 2017 11:19 AM
  • Hi All,

    Microsoft came back to me after they collected diagnostic info from our Server 2016 machine and indeed it was down to an iSCSI issue. We have some legacy connections in the iSCSI Initiator "Targets" that were constantly trying to reconnect. We disconnected them using the "disconnect" button and removed them also from the discovery and favorite tagets tabs also.

    Once this had been done the problems all went away. The same problem was on our Server 2012 R2 Hyper-V machine. Again having removed the legacy connections the problem disappeared. Mirosoft reported they have updated the iSCSI driver in the last few weeks and this is what caused the problem.

    So the solution is to make sure you don't have any iSCSI connection issues.

    Best Wishes,

    Mike

    Thursday, May 25, 2017 1:47 PM
  • Hi All,

    Microsoft came back to me after they collected diagnostic info from our Server 2016 machine and indeed it was down to an iSCSI issue. We have some legacy connections in the iSCSI Initiator "Targets" that were constantly trying to reconnect. We disconnected them using the "disconnect" button and removed them also from the discovery and favorite tagets tabs also.

    Once this had been done the problems all went away. The same problem was on our Server 2012 R2 Hyper-V machine. Again having removed the legacy connections the problem disappeared. Mirosoft reported they have updated the iSCSI driver in the last few weeks and this is what caused the problem.

    So the solution is to make sure you don't have any iSCSI connection issues.

    Best Wishes,

    Mike

    Well, my cause is the same. We removed the legacy iSCSI to solve the problem.
    Thank's for share this information.
    Did you has the report or documentation send by Microsoft?

    Regards,

    Neto



    Friday, May 26, 2017 11:48 AM
  • Thank you so much for this! We are having very similar issues with our Server 2012 R2 servers in our DR data center. We have been trying for weeks to figure out why the servers keep hanging after 2 to 3 days, but could not figure anything out due to the fact that event logs showed nothing and there are no dump files. 

    Looking into your suggestion, we found that the ISCSI initiator for all three servers that are having this issue are pointing to a SAN that is in an entirely different data center. All of this happened after the WannaCry outbreak, where we were up all night patching all 200+ servers in order to protect ourselves. 

    If the problem comes back, so will I, but for now I am pretty convinced this fix will resolve our issue. Thanks again!

    Monday, July 24, 2017 7:53 PM