Click here to monitor SSC
SQLServerCentral is supported by Red Gate Software Ltd.
 
Log in  ::  Register  ::  Not logged in
 
 
 
        
Home       Members    Calendar    Who's On


Add to briefcase ««123»»

AlwaysOn sometimes becomes out of sync Expand / Collapse
Author
Message
Posted Friday, January 31, 2014 4:55 AM
Grasshopper

GrasshopperGrasshopperGrasshopperGrasshopperGrasshopperGrasshopperGrasshopperGrasshopper

Group: General Forum Members
Last Login: Tuesday, September 9, 2014 3:59 AM
Points: 20, Visits: 111
I have upped the session time-out to 15 seconds, i assume this is just a standard ping?

Will see how it goes over the weekend.
Post #1536717
Posted Friday, January 31, 2014 5:05 AM


Ten Centuries

Ten CenturiesTen CenturiesTen CenturiesTen CenturiesTen CenturiesTen CenturiesTen CenturiesTen Centuries

Group: General Forum Members
Last Login: Today @ 7:41 AM
Points: 1,191, Visits: 9,887
michael.mcloughlin (1/31/2014)
I have upped the session time-out to 15 seconds, i assume this is just a standard ping?

Will see how it goes over the weekend.


Careful with that, it really depends what your priority is. Would you rather have your secondary get temporarily out of date, or have your primary hang for 15 seconds when this happens?

Microsoft documentation refers to it as a ping, however I doubt it's an ICMP packet, it'll be a specific heartbeat communication from the Windows Cluster services...
Post #1536721
Posted Friday, January 31, 2014 6:22 AM


SSCertifiable

SSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiable

Group: General Forum Members
Last Login: Today @ 1:08 PM
Points: 6,474, Visits: 13,933
HowardW (1/31/2014)
michael.mcloughlin (1/31/2014)
I have upped the session time-out to 15 seconds, i assume this is just a standard ping?

Will see how it goes over the weekend.


Careful with that, it really depends what your priority is. Would you rather have your secondary get temporarily out of date, or have your primary hang for 15 seconds when this happens?

Microsoft documentation refers to it as a ping, however I doubt it's an ICMP packet, it'll be a specific heartbeat communication from the Windows Cluster services...

Exacto mundo and should ideally have a separate network for the heartbeat traffic. Currently all traffic is pushed down the "same pipe"


-----------------------------------------------------------------------------------------------------------

"Ya can't make an omelette without breaking just a few eggs"
Post #1536748
Posted Friday, January 31, 2014 9:04 AM
Grasshopper

GrasshopperGrasshopperGrasshopperGrasshopperGrasshopperGrasshopperGrasshopperGrasshopper

Group: General Forum Members
Last Login: Tuesday, September 9, 2014 3:59 AM
Points: 20, Visits: 111
Thanks guys, apologies if this sound daft but im the perfect example of the accidental DBA. Having come from a job with no exposure to clulstering and SQL (previous job was in schools) im still working this out in my head.

I have 3 network cards on the VMs all connected to different distributed switches which are tagged on different VLANs. How do i go about using a different network for the heartbeat as i cannot see anything related to this on the existing virtual cluster and physical cluster, only the public and 2 private iSCSI vlans are listed in failover manager.

From what i read when looking into this it is suggested that a heartbeat isn't required any more from SQL 2008 onwards? If this isn't recommended would another network card on each server on the same VLAN configured with a private IP (192.x.x.x for example) be enough for the heartbeat? obviously this would be set to Cluster use only?
Post #1536828
Posted Friday, January 31, 2014 10:40 AM


SSCertifiable

SSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiable

Group: General Forum Members
Last Login: Today @ 1:08 PM
Points: 6,474, Visits: 13,933
michael.mcloughlin (1/31/2014)
From what i read when looking into this it is suggested that a heartbeat isn't required any more from SQL 2008 onwards?

Yes but when running the cluster validation the report will still bleat about not having a separate network!


michael.mcloughlin (1/31/2014)
If this isn't recommended would another network card on each server on the same VLAN configured with a private IP (192.x.x.x for example) be enough for the heartbeat? obviously this would be set to Cluster use only?

A separate network would be advisable as you only have 1 Nic available


-----------------------------------------------------------------------------------------------------------

"Ya can't make an omelette without breaking just a few eggs"
Post #1536889
Posted Monday, February 3, 2014 3:30 AM
Grasshopper

GrasshopperGrasshopperGrasshopperGrasshopperGrasshopperGrasshopperGrasshopperGrasshopper

Group: General Forum Members
Last Login: Tuesday, September 9, 2014 3:59 AM
Points: 20, Visits: 111
OK im getting mixed messages on whether a heartbeat network is needed, Microsoft recommend you dont use one for 2008 onwards as it can have adverse effects. I can get another VLAN created and add another NIC to the servers but if Microsoft say it isnt needed then why bother? Surely a 10GB interface should be able to handle a simple heartbeat between nodes?
Post #1537235
Posted Monday, February 3, 2014 7:44 AM


SSCertifiable

SSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiable

Group: General Forum Members
Last Login: Today @ 1:08 PM
Points: 6,474, Visits: 13,933
michael.mcloughlin (2/3/2014)
OK im getting mixed messages on whether a heartbeat network is needed, Microsoft recommend you dont use one for 2008 onwards as it can have adverse effects.

A lot changed in Windows 2008, Teamed adapter support was better as well as the improvements to the heartbeat networking. This post gives excellent detail of the changes.
The detail in the link above mentions the special virtual heartbeat adapter which uses an APIP address, thing is it will have to bind itself to a NIC on the server to be able to complete the heartbeat communication.


michael.mcloughlin (2/3/2014)
Surely a 10GB interface should be able to handle a simple heartbeat between nodes?

Yes but this 10Gbps connection is shared by however many VMs are connecting through the vSwitch, or are you saying that this VM is the only VM communicating through that switch?


-----------------------------------------------------------------------------------------------------------

"Ya can't make an omelette without breaking just a few eggs"
Post #1537327
Posted Monday, February 3, 2014 7:48 AM
Grasshopper

GrasshopperGrasshopperGrasshopperGrasshopperGrasshopperGrasshopperGrasshopperGrasshopper

Group: General Forum Members
Last Login: Tuesday, September 9, 2014 3:59 AM
Points: 20, Visits: 111
Currently there are only these nodes with 10GBPs interfaces on this vSwitch. Any other VMs on the same vSwitch have 1Gbps nics configured. I will take a look at the URl you gave later, currently sorting a failed disk in the san
Post #1537329
Posted Tuesday, February 4, 2014 3:57 AM


SSCertifiable

SSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiableSSCertifiable

Group: General Forum Members
Last Login: Today @ 1:08 PM
Points: 6,474, Visits: 13,933
what version ESX\VSphere you using?

-----------------------------------------------------------------------------------------------------------

"Ya can't make an omelette without breaking just a few eggs"
Post #1537672
Posted Tuesday, February 4, 2014 8:50 AM
Forum Newbie

Forum NewbieForum NewbieForum NewbieForum NewbieForum NewbieForum NewbieForum NewbieForum Newbie

Group: General Forum Members
Last Login: Monday, September 22, 2014 2:14 AM
Points: 6, Visits: 93
Hi gurus !!

Check if you have any error in the SQL Server Errologs related with avaliable workers and raise the value in case you find this error.

How many databases you have using AlwaysON ? I plan tor prepare an 5000 databases AlwaysOn setup and my first tests ended with a bad result due a lack of threads. I increased the maximum workers value and it didn't help. Any of you have any experience with this amount of databases ? My understand is that it won't work with the hardware I have avaliable ( Two 16 corees 65 Gb of RAM ).

Check this table (The number of workers might vary depending the number of cores you have).

http://msdn.microsoft.com/en-us/library/ms190219.aspx



Post #1537810
« Prev Topic | Next Topic »

Add to briefcase ««123»»

Permissions Expand / Collapse