I used DataDog. At the time, I didn't have any experience with python. I had to depend the company to make patches to things that didn't work. And anything custom, I just did a cron job instead of trying to integrate with their tool.
They marketed themselves as a monitoring company but they really collect a boatload of performance info that you can wade though to find out what's going on on the system being monitored. The alert setup and other monitor type things have to be done through the web GUI. I don't think you can code it though an API. You put whatever instrumentation you want on the server to be monitored, then clone those files and any local changes you made to their code. It doesn't play well for setting up as an automated solution like stand alone monitoring packages. It's $15/node, which, if you 5000 nodes, you're talking real money. Go with an on-prem solution rather than something that's cloud based. Those can be configured and deployed painlessly. -- Michael Vilain 650-322-6755 > On 26-Jul-2018, at 10:05 AM 🌞, Malcolm Greene <pyt...@bdurham.com> wrote: > > Looking for feedback on anyone who's using a cloud based > monitoring/logging service with Python's standard lib logging module, > eg. services such as Datadog, Loggly, Papertrailapp, Scalyr, LogDNA, > Logz.io, Logentries, Loggr, Logstats? > Would appreciate hearing your experience, pros, cons, > recommendations, etc. > Thanks! > Malcolm > -- > https://mail.python.org/mailman/listinfo/python-list -- https://mail.python.org/mailman/listinfo/python-list
