Managing Chrome Updates is a nightmare?
It appears that managing updates to Google Chrome is a complete nightmare when working with TestComplete. I am curious how others manage this. This week Google pushed out Chrome 42 and all my PCs...
Forum Discussion
To clarify, I am not opposed to keeping Chrome updated or TestComplete. I just want to manage this on my timeframe since it is not a small undertaking when you are using 42 PCs.
My biggest fear is that Chrome will push out an update the day before I need to do final testing on one of our major releases. We would have to delay our software release or skip Chrome testing and hope testing Firefox and IE is good enough. Even if TestComplete has a patch available the day the Chrome update comes out, it would still mess me up because I can't update 42 PCs in an hour.
I simply want to be able to say "wait until next week to update Chrome" so I can do it at a timeframe that does not disrupt my business.
I get it, but you're really going to have 42 PCs with an outdated browser?
Or did I misunderstand this? "update 42 PCs"
"Firefox and IE is good enough."
This is ultimately what our team decided...to stick with just IE for automated testing until this process is improved upon by SmartBear. Ultimately the decision was made when we reevaluated the purpose of automated testing (good to to do every now and then for sanity sake) and it was determined that the ultimate goal is to catch major issues with a release. Testing every browser and every scenario was not determined to be a focal point of automation since we have manual testers running those browsers. So far, for us, this has proven to be time better spent and we're able to dedicate more time to automated script expansion.
You said:
This is ultimately what our team decided...to stick with just IE for automated testing until this process is improved upon by SmartBear. Ultimately the decision was made when we reevaluated the purpose of automated testing (good to to do every now and then for sanity sake) and it was determined that the ultimate goal is to catch major issues with a release. Testing every browser and every scenario was not determined to be a focal point of automation since we have manual testers running those browsers. So far, for us, this has proven to be time better spent and we're able to dedicate more time to automated script
Fair point and this might be what we are ultimately stuck doing. We don't test everything on all browsers but we do have certain types of regression testing that we want to run on each release and on each currently supported major browser.
-" we do have certain types of regression testing that we want to run on each release"
Understood.
For us it was a nice idea to run on every browser, and sounded good to management, but we could not justify the logic behind it.
Not sure I understand this question?
I get it, but you're really going to have 42 PCs with an outdated browser?
Or did I misunderstand this? "update 42 PCs"
We have 42 PCs. All of them currently have Chrome 42 installed, TestComplete 10.60.3387.7 and TestExecute 10.60.3387.11. If Google were to push out version 43 of Chrome, then all of my PCs would have an out of date version of Chrome. Even if TestComplete has a patch for Chrome 43, it seems they only create this patch for the latest version of TestComplete. So, if the latest version of TestComplete/TestExecute happens to be 10.70...... then I would be forced to do the following to be able to test on Chrome again:
- Update TestComplete to version 10.70 on all 42 PCs
- Update TestExecute to version 10.70 on all 42 PCs
- Apply the patch to TestComplete on all 42 PCs
- Apply the patch to TestExecute on all 42 Pcs
This process takes most of a day to complete. I simply want to be able to tell Chrome to wait until next week to update so I have time to complete steps 1-4 first.