Provide visibility within BBCRM of system performance metrics
We often experience periods of slowness in various parts of the system. It would be really useful to be able to look at a page in BBCRM, or a page on the BB web site, specific to our instance, which would tell us how the system is performing. That way, when our users complained that the system is slow, e.g., Data Lists timing out, Pages being slow to load, we could look at some metrics to see whether there is a system problem.
In https://kb.blackbaud.com/knowledgebase/articles/Article/51454
you suggest some tools, SQLSnapshot and FiddlerCaps, for use on self-hosted environments. Could one of those be made available for our use in BB-hosted environments?
-
David Wanless
- Dec 14 2022
- Reviewed: Voting Open
| Organization Name (Please enter full organization name) | The Wilderness Society Ltd |
| Reported Version | 4.0 |
Related ideas
Agreed! We are hosted and have experienced poor performance as well after going to Azure. Some days are good and other days, it is really bad. Just like you we have been working with Blackbaud for months now trying to figure out what works best for us (fiddling with compatibility levels and cardinality estimator). We do have access to SQL snaps that shows us what is running however we do have to depend on Blackbaud to kill a process that we see running in SQL snaps. It would be beneficial to have a way to kill a slow/blocking process rather than depending on Blackbaud to do so. It is very unfortunate Blackbaud does not have answers to the slowness/poor performance of the system.
Attachments Open full size
agreed
Attachments Open full size
We know that one of the main causes of poor performance is record locking, so it would be really useful to have visibility of which processes were involved in record locking, in real time, and looking backwards. When Blackbaud support has been able to investigate the logs and identify which processes were involved this has been very helpful in identifying problems so that we can adjust or reschedule processes. Unfortunately, getting this information from BB support takes hours to weeks so it is unable to help at the time that the actual slowdown is happening, so we just have to grin and bear it, and apologise to our users. If we could see what was happening at the time, that would be very useful as we might be able to stop whatever was implicated in the problem.
Attachments Open full size
Agree, would be very useful. The status page is fine for overall high level performance but something which identifies individual systems would be brilliant.
Attachments Open full size
I agree that this would be extremely useful for transparency and efficiency 5/5
Attachments Open full size