[bisq-network/roles] Monitoring Operator/Maintainer (#10)

[Harry MacFinned]

### Cycle 3 (June/July) Report - Prices monitoring

**Timeouts**
- some timeouts on my monitoring side spoil the statistics, but excluding those observer-side timeouts, the overall statistics seem in the usual range.
* Prices are given from the same onions as fees -> see the weekly reports (above) on fees for weekly/daily details. See https://github.com/HarryMacfinned/bisq-monitor for detailed data.

**Divergence between nodes**
* 4 issues this cycle. All corrected rapidly.
* apart above there were mainly only punctual alerts, ie false alarms (and also only on secondary assets)

**Development**
- I've added the monitoring of the timestamps to the general monitoring. This should help to catch the deviations due to data flow being stalled. Currently the alert threshold is fixed at 600 seconds (both inter-nodes and intra-nodes measurement)
The timestamp monitoring is very recent, there is however some stats :
- for 1000 pings, 10 alerts were triggered = 1%
- the issues were in a row (2019-07-08 14:05 - 2019-07-08 14:55) and concerned simultaneously all the nodes (which strongly suggests that the issue was on the provider side).

- The threshold to alert for prices deviations has also be changed from 5% to 2%. This adds some alerts for very small prices (10 power-7 and so, eg DOGE, SC), but apart that it seems ok.

All the alerts are given on the slack #bisq-monitor and #bisq-pricenode channels.

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/bisq-network/roles/issues/10#issuecomment-509290795
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bisq.network/pipermail/bisq-github/attachments/20190708/d6f4572d/attachment.html>