I was wondering if anyone would be able to help me out with enabling logging on Telegraf. I see that there is a logrotate file that points to /var/log/telegraf/telegraf.log but I am not able to see how one enables logging to this location
I have tried with the -debug flag but no log file was generated
Mattch has quit
ldumont has quit
gunnar has quit
gunnar joined the channel
gunnar has quit
gunnar___ joined the channel
tuupola
iirc quiet = false in the telegraf.conf
might remember wrong though
athenot joined the channel
ghounds joined the channel
ldumont joined the channel
heydrick joined the channel
ldumont has quit
gunnar___ has quit
Gen joined the channel
makaveli0227TL joined the channel
gunnar joined the channel
gunnar has quit
gunnar joined the channel
areski joined the channel
gunnar has quit
gunnar joined the channel
gunnar has quit
gunnar joined the channel
gunnar___ joined the channel
gunna____ joined the channel
gunnar has quit
GiLgAmEzH joined the channel
gunnar___ has quit
radiocats joined the channel
radiocats has quit
gunna____ has quit
gunnar joined the channel
radiocats joined the channel
gunnar___ joined the channel
gunnar has quit
Gen joined the channel
gunnar joined the channel
gunna____ joined the channel
gunnar___ has quit
szhem joined the channel
gunnar has quit
gunnar joined the channel
gunnar___ joined the channel
gunna____ has quit
gunnar has quit
gunnar___ has quit
gunnar joined the channel
gunnar___ joined the channel
gunna____ joined the channel
gunnar has quit
gunnar___ has quit
gunnar joined the channel
gunna____ has quit
awallin__
is there anything available for parsing a ResultSet (looks like JSON) returned by a python query?
I would want time as python datetime and my fields are floats
gunnar___ joined the channel
heydrick joined the channel
gunnar has quit
gunnar___ has quit
gunnar joined the channel
GiLgAmEzH has quit
jjmalina has quit
GiLgAmEzH joined the channel
productiondba joined the channel
gunnar___ joined the channel
gunna____ joined the channel
gunnar has quit
gunnar___ has quit
gunna____ has quit
gunnar joined the channel
gunnar___ joined the channel
heydrick1 joined the channel
gunnar has quit
jjmalina joined the channel
with 'top', is it normal to see about 18 influxdb processes running? the machine has 24 cores.
heydrick joined the channel
Gen joined the channel
gunnar___ has quit
gunnar joined the channel
radiocats joined the channel
radiocats joined the channel
gunnar has quit
gunnar___ joined the channel
heydrick has quit
makaveli0227TL has quit
gunnar___
awallin__: are you using top or htop?
gunnar___ has quit
gunnar joined the channel
gunnar___ joined the channel
gunnar has quit
nicksellen has quit
gunnar joined the channel
gunna____ joined the channel
gunnar___ has quit
gunnar___ joined the channel
gunnar has quit
gunna____ has quit
gunnar joined the channel
gunna____ joined the channel
gunnar___ has quit
gunnar has quit
gunna____ has quit
gunnar joined the channel
gunnar___ joined the channel
gunnar has quit
thumpba joined the channel
gunnar___ has quit
gunnar joined the channel
gunnar___ joined the channel
gunnar has quit
gunnar joined the channel
productiondba joined the channel
gunnar___ has quit
heydrick joined the channel
gunnar has quit
gunnar___ joined the channel
harobed joined the channel
harobed3 joined the channel
dancleinmark joined the channel
jafo
gunnar___: (Back to my problems with network traffic graphs) I have had a lot of problems with managing anything that has a number that grows over time, like network bytes/pkts in/out, I think disk IOs is another one. I've tried the derivitive functions but I just don't seem to be able to figure out how to put all the pieces together to make something that makes useful graphs.
gunnar___: I've tried cobbling together various examples I've seen in issues and the like, but still really struggle with it. Maybe it's more something that could be addressed with Grafana for a better UX, but just having something that works is where I'm stuck now.
Happily, the one problem I had where I couldn't get large spikes at the beginning and end of my samples to go away seems to have been fixed. I think that was a derivitive problem in influx with the first/last sample getting the full value.
Anyone know of any samples for how to graph telegraf network traffic?
anettu
maybe you're graphing the network traffic from a multiple hosts and thatswhy it doesn't make sense?
harobed joined the channel
harobed3 joined the channel
gunnar joined the channel
jafo
anettu: Hmm. Can't just get an overall network rate? Like the derivitive is applied after the sum, so wrap arounds freak it out or something?
gunnar___ has quit
anettu: I'm also having problems if I group by host. Right now my graphs are dominated by huge spikes that don't make any sense.
heydrick joined the channel
anettu
have you tried with fill(none)?
jafo
anettu: It has "fill(null)" in it.
SELECT derivative(last("bytes_recv"), $interval) FROM "net" WHERE $timeFilter GROUP BY time($interval), "host" fill(null)
anettu
yeah, exactly the same one works for me
jafo
anettu: Are you setting a "group by time interval"?