The problem is, that the error message is logged but the pipeline considers the task successful, so there is no warning that anything has gone wrong. It would be more helpful if the task would error in such a way that the pipeline could know about it.
Appendix: full log of the run which had the time-out failure:
2021-09-22T19:47:33.8079932Z ##[section]Starting: Tag Build If Release
2021-09-22T19:47:33.8212932Z ==============================================================================
2021-09-22T19:47:33.8213283Z Task : Tag Build or Release
2021-09-22T19:47:33.8213545Z Description : Adds tags to a build or release
2021-09-22T19:47:33.8213797Z Version : 0.1.114
2021-09-22T19:47:33.8214047Z Author : Colin Dembovsky (colinsalmcorner.com)
2021-09-22T19:47:33.8214468Z Help : More Information
2021-09-22T19:47:33.8214876Z ==============================================================================
2021-09-22T19:47:55.4536789Z (node:5260) UnhandledPromiseRejectionWarning: Unhandled promise rejection (rejection id: 1): Error: connect ETIMEDOUT 13.107.42.20:443
2021-09-22T19:47:55.4642539Z ##[section]Finishing: Tag Build If Release
Very occasionally, the TagBuild task will fail with a timeout error like this:
The problem is, that the error message is logged but the pipeline considers the task successful, so there is no warning that anything has gone wrong. It would be more helpful if the task would error in such a way that the pipeline could know about it.
Appendix: full log of the run which had the time-out failure: