Add new panel - cache hit ratio

[pchovelon]

Add a new panel to follow cache hit ratio of databases.
The panel used this promql query :

[sxd]

Hi @itay-grudev @jsilvela
Can you take a look into this ?

[pchovelon]

As @sxd suggested, it could be nice to remove template databases. I'll have a look into it.

[itay-grudev]

Until now I've always added a custom monitoring query for cache hit ratios. I didn't realise we've had the data already.

[jsilvela]

tested out and looks good

[jsilvela]

Just a comment that on testing, I saw the warning notice which complains about metric names.
Pretty annoying "feature" coming from grafana.
See grafana/grafana-prometheus-datasource#159
:/

[pchovelon]

@jsilvela @itay-grudev I just changed the query to exclude template0 and template1 series.

[jsilvela]

@pchovelon not working properly. Malformed JSON. Going to see if I can do a quick fix for you.

[jsilvela]

Ah, yes, it's all the unescaped quotes in the JSON...

[jsilvela]

unescaped quotes in JSON. I left you a suggestion that fixes it, I hope

[pchovelon]

Anything I can do to see this PR merged ? @jsilvela @itay-grudev
Thx :)

[jsilvela]

@pchovelon did you even read my review from 3 weeks ago? The code as it stands doesn't work.

[pchovelon]

@pchovelon did you even read my review from 3 weeks ago? The code as it stands doesn't work.

Oh I see ... I wrote a review ... but it seems you never get it. You should saw it now. Is it right ? 🤔

[jsilvela]

@pchovelon ok I see it now, but I'm confused. The only difference I see between your fix and mine is that you have two newlines. I'll have to test this ... later, possibly today can't promise.

[jsilvela]

For the sake of readability of this thread (tsk, Github).
Your proposed edit will work, so I'll approve after you apply @pchovelon