Sharepoint Performance issue Datacenter Amsterdam

[Microsoftie]

Our organisation is hit by Performance Problems with SharePoint Online since about january 2019. All 2-3 weeks we experience performance issues on our tennant in Amsterdam for about one to two hours. Then suddenly it dissapears and all is back to normal. We checked everything on our side, we have same problems outside the company as well, so its impossible that we or our Network is the root cause of this. The technician from MS said once they had a peak in the infrastructure, but did not say any more. We had several tickets with microsoft, but incidents are closed right after everything is working again. But its a general problem nobody at MS wants to recognise. Yesterday they had a «Global» issue with  Sharepoint and fixed it overnight. Today we informed our 10000 users that everything is back ok, and  a half a hour later we are hit with performance issues again. Are we the only ones having issues like this? 

[#2]

I'm not seeing those problems but I have a customer that uses heavil SPO and they have experiencing from time to time performance problems

[#3]

Exactly this it what i mean, from time to time performance issues that nobody can explain, is it Amsterdam Datacenter they are using? @Juan Carlos González Martín 

[#4]

Hi @Microsoftie ,

From time to time we have had issues but nothing ongoing. 

In some instances we have had issues that have had to be fixed in the backend after a support call. There are a number of checks that I normally do which you may have done already. 

1. Check your ping to the data centres - https://azurespeedtest.azurewebsites.net/ . Just proves its not your network. 

2. In any page response there is a SharePointHealthScore in the Response header. 0 is good, 9 is very bad indicated load or the service is not capable of returning the pages. 

3. Check Request duration, this is how long it takes to process the page in milliseconds. 

Might help with the conversation with Support. 

I once had a SharePoint Online tenant that stopped responding at 11.43 every morning, we could sit there and time it. Turned out another customer was running some sort of timer job at a time that was out of hours to them but not to my customer. 

Andy

[#5]

@Andrew Hodges Thanks, we have had latency issues the exact same time too for a while, every monday between 11:00 and 12:00 AM, then it switched over to tuesdays....about the same time. Now its any day of the week...and there is no pattern anymore. 

Sharepoint response times got better, single pageload was between 1min 30sec and 2minutes or just forever. Now its under a minute when we have a incident. This is far off from normal operation, but only for a hour or so and also unregular with no pattern. And that makes it unmeasurable an risky. After every incident i thought, this was the last one and its over. But today proved again that i was wrong. My guess is that Amsterdam as a tennant is getting unusable for us in the state it presented itself this year for us. Maybe its just overloaded with old hardware or too much customers. The risk of latency issues from 3rd parties seems very high (i had this in my mind now for months). Thats why i created the thread, to hear if others have latency problems too in Amsterdam. 

Thank you Andrew

Oh it was tuesday today....