Monitor disk space used by plain text for queue timings data

[makyen]

With the merge of PRs #3971, PR #6705, and #6706, SmokeDetector is now storing queue timing information in text format in bodyfetcherQueueTimings.txt. While this should substantially reduce the amount of total amount of data written^a, it is expected to notably increase the amount of disk space required for that data (text which includes the domain name for each numeric data point vs a binary format which doesn't duplicate the domain name for each number entry and which has the floating point number in binary rather than text).

We should monitor the amount of space actually occupied over time by the text formatted data. If it does get large, then we should consider options for dealing with it. Off the top of my head, some possibilities for dealing with it include:

1. periodically adding the text formatted data to the binary (pickle) format data and then clear the file containing the text data. [queue_timings.py has code which will read both files into a single data structure.]
2. copy/move it elsewhere.
3. delete it
4. something else?

---

a. The binary (pickle) format writes more data because the entire amount of data was being written to the pickle file after every new data point. The new code bot just appends the data to the text file, thus not rewriting the data, and only writes to the file every 128 data changes (or upon reboot/shutdown).

[stale]

This issue has been closed because it has had no recent activity. If this is still important, please add another comment and find someone with write permissions to reopen the issue. Thank you for your contributions.