time on server - clock loose 20 sec every min
Hi :) I have two physical SQL 2008 R2 cluster nodes on 2008 R2 SP1. On one node only I have problem with time / clock. Aprox. every minute clock loose 20 sec. Nodes are configured to sync with DOMHIER, ie. DC. Second node does not have time problems. I've made changes in w32time service, and put SpecialPollInterval for every 120 sec. If I put command w32tm /resync sometimes nothin's happen and sometimes time changes to be same as in DC but as I wrote after 1 minute clock loose 20 sec... and time skew from 4 to 8 minutes. I will try to move instance to another node and reboot node to check time in BIOS but right now I can't figure why clock is slowing... :(
September 1st, 2012 4:18pm

Is the CMOS battery on the mainboard good? Have you tried replacing it? If you can have some downtime on one node boot into the BIOS and track the time to see if the problem occurs w/o the OS loaded. This way you can determine if the problem is hardware or not. Miguel Fra | Falcon IT Services, Miami, FL www.falconitservices.com | www.falconits.com | Blog
Free Windows Admin Tool Kit Click here and download it now
September 1st, 2012 4:20pm

Thank's for reply Falcon :) As I wrote, I will try early in the morning to move instance of SQL to another node and reboot server to see in BIOS what is going' on...
September 1st, 2012 4:27pm

After moving instance, restarting node and entering into bios I think that this is hardware related problem. Clock in bios is ok but if I try to move trough BIOS options using key arrowrs it seems BIOS is freezing / locking up. I can move trough some option but than all freeze... :(
Free Windows Admin Tool Kit Click here and download it now
September 2nd, 2012 11:31pm

When I run command: w32tm /query /status /verbose I've notice in output difference between two servers: first node with time problem: Leap Indicator: 0(no warning) Stratum: 7 (secondary reference - syncd by (S)NTP) Precision: -6 (15.625ms per tick) Root Delay: 0.1737061s Root Dispersion: 8.1336780s ReferenceId: 0x0A003604 (source IP: 11.12.13.14) Last Successful Sync Time: 3.9.2012 16:43:17 Source: dc1.aaa.bbb Poll Interval: 10 (1024s) Phase Offset: 6.9053454s ClockRate: 0.0156001s State Machine: 1 (Hold) Time Source Flags: 0 (None) Server Role: 0 (None) Last Sync Error: 0 (The command completed successfully.) Time since Last Good Sync Time: 523.9862143s and second one which is good: Leap Indicator: 0(no warning) Stratum: 7 (secondary reference - syncd by (S)NTP) Precision: -6 (15.625ms per tick) Root Delay: 0.1748810s Root Dispersion: 4.4958856s ReferenceId: 0x0A003604 (source IP: 11.12.13.14) Last Successful Sync Time: 3.9.2012 14:39:12 Source: DC1.aaa.bbb Poll Interval: 15 (32768s) Phase Offset: 0.0012329s ClockRate: 0.0156009s State Machine: 2 (Sync) Time Source Flags: 2 (Authenticated ) Server Role: 0 (None) Last Sync Error: 0 (The command completed successfully.) Time since Last Good Sync Time: 8200.8473781s Does anyone knows what is difference between: State Machine: 1 (Hold) State Machine: 2 (Sync) and is this some kind of final state or is this some setting which I can change ? And why time source flags are different: Time Source Flags: 0 (None) vs Time Source Flags: 2 (Authenticated ) W32Time remote protocol specification: 2.2.7 State of Time Service The state of the time service represents the current state of the time service in the clock discipline algorithm. For more information on clock discipline algorithms, see [NTP-TR9733i] and [NTP-TR9733] section 4.3. The clock discipline algorithm is optional and implementation-specific.<9> The following table lists the values that SHOULD be supported. Value Meaning UNSET 0x00000000 The clock is not synchronized. HOLD 0x00000001 The clock is adjusting the time difference only, not the clock rate. SYNC 0x00000002 The clock is synchronized. Both the time difference and the clock rate are adjusted. SPIKE 0x00000003 A spike was detected, but the clock discipline algorithm cannot determine if the clock is no longer synchronized or if the spike was just a network jitter.
September 3rd, 2012 11:23am

Hi, Generally, the third party security software, such as antivirus program and firewall may cause such issue. You may disable them to check the result. If it does not work, I also would like to suggest you try the Microsoft Fix it tool to reconfigure an authoritative time server from the following link: http://support.microsoft.com/kb/816042 Regards, Arthur Li TechNet Subscriber Support If you are TechNet Subscription user and have any feedback on our support quality, please send your feedback here.Arthur Li TechNet Community Support
Free Windows Admin Tool Kit Click here and download it now
September 3rd, 2012 10:52pm

Hi, Generally, the third party security software, such as antivirus program and firewall may cause such issue. You may disable them to check the result. If it does not work, I also would like to suggest you try the Microsoft Fix it tool to reconfigure an authoritative time server from the following link: http://support.microsoft.com/kb/816042 Regards, Arthur Li TechNet Subscriber Support If you are TechNet Subscription user and have any feedback on our support quality, please send your feedback here.Arthur Li TechNet Community Support
September 3rd, 2012 10:54pm

Hi, I would like to confirm what is the current situation? If there is anything that I can do for you, please do not hesitate to let me know, and I will be happy to help. Regards, Arthur Li TechNet Subscriber Support If you are TechNet Subscription user and have any feedback on our support quality, please send your feedback here.Arthur Li TechNet Community Support
Free Windows Admin Tool Kit Click here and download it now
September 9th, 2012 11:17pm

This topic is archived. No further replies will be accepted.

Other recent topics Other recent topics