Using the new interface (not legacy), the dashboard graphs seem pretty flaky for me. Is anyone else seeing issues with it? Any ideas what I can do to troubleshoot?
Specific issues are:
1. The HDD graphs only appear sometimes. I opened two tabs, about 2 minutes apart from each other, both pointed at the dashboard, and tab 1 has no HDD graphs, tab 2 does.
2. The system information object never seems to update (there isn't much to update there to be fair, just the time and uptime).
3. The HDD object seems to very rarely show a temperature for the drives (and the temperature is available via smartctl, and directly on the reports page, although, weirdly, not in netdata) - EDIT: Actually I'm not convinced the report page shows it consistently either. I just loaded it up with all disks and only got temps for 4 out of 6 drives.
4. Not really a bug, but the update rate is glacially slow. Is this tunable? Note that I do not mean the time to load the graph, I mean the interval where data is collected and displayed.
5. The update rate for each graph appears to be different! 3 minutes for memory usage, CPU temperature has a weird cycle, can be either 20 or 30seconds, and memory usage, cpu temp and load average seem to be staggered by 1 minute. None of these metrics should be very expensive to gather at a rate of seconds rather than minutes
None of these things are critical, but if I can help by providing more information...
EDIT: I'm new here, but it sounds like the new UI is, well, new in 11.2, so apologies if the above is known already. I searched and found several issues with the dashboard, although not these specific ones.
Specific issues are:
1. The HDD graphs only appear sometimes. I opened two tabs, about 2 minutes apart from each other, both pointed at the dashboard, and tab 1 has no HDD graphs, tab 2 does.
2. The system information object never seems to update (there isn't much to update there to be fair, just the time and uptime).
3. The HDD object seems to very rarely show a temperature for the drives (and the temperature is available via smartctl, and directly on the reports page, although, weirdly, not in netdata) - EDIT: Actually I'm not convinced the report page shows it consistently either. I just loaded it up with all disks and only got temps for 4 out of 6 drives.
4. Not really a bug, but the update rate is glacially slow. Is this tunable? Note that I do not mean the time to load the graph, I mean the interval where data is collected and displayed.
5. The update rate for each graph appears to be different! 3 minutes for memory usage, CPU temperature has a weird cycle, can be either 20 or 30seconds, and memory usage, cpu temp and load average seem to be staggered by 1 minute. None of these metrics should be very expensive to gather at a rate of seconds rather than minutes
None of these things are critical, but if I can help by providing more information...
EDIT: I'm new here, but it sounds like the new UI is, well, new in 11.2, so apologies if the above is known already. I searched and found several issues with the dashboard, although not these specific ones.
Last edited:
