CA-390129: Handle & log exceptions in dump_xapi_rrds(hidden and disabled)

[bernhardkaindl]

CA-390129: Handle & log exceptions in dump_xapi_rrds() from xen-bugtool --entries=persistent-stats:

While this capability is hidden, is executed when xen-bugtool -y -a is called.

[codecov]

Codecov Report

Merging #94 (b468eb5) into master (d0e569d) will increase coverage by 2%.
Report is 11 commits behind head on master.
The diff coverage is 100%.

Additional details and impacted files

@@          Coverage Diff          @@
##           master    #94   +/-   ##
=====================================
+ Coverage      88%    91%   +2%     
=====================================
  Files          18     19    +1     
  Lines        2196   2184   -12     
=====================================
+ Hits         1939   1993   +54     
+ Misses        257    191   -66     

Files	Coverage Δ
xen-bugtool	88% <100%> (+3%)	⬆️

Flag	Coverage Δ
python2.7	89% <100%> (+2%)	⬆️
python3.10.13	90% <100%> (+3%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

[coveralls]

Pull Request Test Coverage Report for Build 8378917082

Details

• 8 of 8 (100.0%) changed or added relevant lines in 1 file are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.1%) to 92.641%

---

Totals
Change from base Build 8378558900:	0.1%
Covered Lines:	2115
Relevant Lines:	2283

---

💛 - Coveralls

[ashwin9390]

Could we use fstring here instead of string formatter log(f"{e}, {traceback.format_exc()}")

[bernhardkaindl]

Could we use fstring here instead of string formatter log(f"{e}, {traceback.format_exc()}")

Thanks for the approval! I'm intentionally not using f-strings in xen-bugtool itself yet:

• Some of the recent fixes are also be relevant for Yangtze:
  • In case we want to backport such fixes with the next hotfix to Yangtze, we can use the already reviewed, tested and proven code (less risk of messing up the backport).
• XS8 is running xen-bugtool using the Python2.7 runtime currently:
  • Rationale 1: Stay clear of any missing Py3 updates.
  • Rationale 2: Be a staging and testing ground for more mission-critical Yangtze backports. (refers to the backport story)

Of course this will change at least when Yangtze support would end.