Quote from: papampi on November 06, 2017, 11:45:50 AM

Quote from: Tigel on November 06, 2017, 10:39:10 AM

Just installed this nvoc on one of my rigs, looks good. So questions though (no, I did not read all the pages..):

- Is there a way to limit telegram messages to be sent only when there is something wrong, i.e. reboot of server or miner is needed ?
- My rig has some stability issues, I am mining Ethereum with ethminer. In order to track the GPUs causing problems they would have to be run as separate instances/sessions, (xmos does this by default). Any plans for implementing such feature? If not maybe I try with claymore or plug in xmos USB stick, adjust overclock settings and copy them to nvOC.

--
Tigel

You can set telegram to no, but may lose some info, or set to high times like 1440 (once a day)
Papampi telegram will send alerts either way

I've implemented an idea where i've reduced the number of telegram notifications by integrating with https://healthchecks.io/checks/
I have a crontab job run every 5 minutes where I curl the healthchecks.io checks:
*/5 * * * * curl -fsS --retry 3 https://hchk.io/CHECK_ID > /dev/null

And in there healthchecks.io/checks you would configure how often you would expect to be notified and when you would raise an issue. There is an integration with Telegram, where the bot can be configured to send you an message in case of ping not received during the grace period.

And then the Bali_TELEGRAM_ALERTS can be scheduled to not send messages at all, only when WATCHDOG seems appropriate.

at10ti0n

Posts

Merits