Share
Report
12Hello,
Our company has been experiencing frequent slowdowns to the point we have had customers walk out and leave and employees taking hours to process a single sales and or purchase order. The first two days after go live were unbearable, microsoft said there was a memory issue and moved us to another isolated server, this stabilized us for about a two weeks, now we have had two more days of down time unable to process orders. We have been working with our partner and it seems Microsoft may not be giving us large amounts of information to try to pinpoint what's going on, its hard to say as I dont know all of the info our partner has been provided. We have an integration service (rapidi) that is reading and injecting customer and jobs into the system from salesforce, they are pointing towards this as the problem but the api calls do not seem excessive, below is the latest response from Microsoft. We turned off the web services to block the integration and over a 2-3 hour span did not recover performance, so we were ruling out the integration. We then turned it off last night and reenabled closer to 930, by 11-12 users were complaning of slowdowns. We also use Jet Reports and were running some historical aged AR reports that took a good 20 minutes to run in Jet. Does anyone have experience with performance issues with BC that have severely impacted business and operations? I would think resources would be allocated as needed, but we are frequently hitting massive slowdowns and lag times for multiple offices and remote users.
Also, I set up an app insight on azure and linked the instrumentation key with our environment but am not getting any data showing on our insight server. Is there any additional setups or anything else to help get data flowing. I am going to logs and running, Traces | Take 100. Any suggestions would be greatly appreciated.
-----
From Microsoft
I did find an interesting query that according to the monitor was probably a significant part of the performance issues over the last couple of days. This might correspond to the RapidCustomer OData processing . There is a query on the Customer table that is returning all records past a certain value (I can’t see the values, so not sure if this is the first or last or somewhere in the middle). The table below shows how many times it ran and for what time frame.
It does look like you have shutdown the majority of OData and SOAP calls. They are now the lowest consumer of WS on this service.
I have also found a query against the Customer table that has 66 FlowFields with only 38 having VSIFT indexes to support them. This query could have a pretty good size impact depending on the filtering and counts on the supporting tables. With Azure SQL, it is very important that each query is supported by a key.
------
All responses (
Answers (
