Error: Disk can't keep up

[Gunner]

Well... It seems to me that we are either suffering from mass delusions Simultaneous but independent drive issues Or something changed in one of the latest updates

[butbright]

I am running BI in a VM on proxmox solved this issue by giving the VM 8cores and 16GiB of ram.

At first I was running 6 cores and 8GiB ram with a ZFS raid with r/w in excess of 600mbs and got the error. I changed a bunch of setting with the cluster and then I thought, let me just pass through a gen4 NVME raid with 10k r/w speeds. I still received the error, and was tempted to reimage everything when I noticed it happened when it was saving clips and updating alerts, leading me to believe it needed more cores or ram. Haven't had an issue since adding these.

[Gunner]

Well, I don't know what is going on... My C: is an SSD and I have spinning rust for a D: (dedicated to BI)
After many adjustments and still getting errors, I started from scratch and took the spinning rust out of the equation... But within a few hours, I am still getting some errors - Clip: Disk can't keep up... I mean, aside from the obvious "something slow sometimes", what does that error exactly refer to anyhow???

I ran the speed test... But I dunno if this is good, bad or ugly

Capture.PNG (29.38 KiB) Viewed 340 times

[mdw]

With the latest build, I am having a lot less alerts, but it is still occurring every few days.
Running Dahua cameras with Intel hardware acceleration.

Code: Select all

Clips: 1708 items, 5500/5502GB; D: +161.4GB, C: +747.4GB

5/5/2024 12:17:38 PM Front: Clip: Disk can't keep up
5/5/2024 12:14:04 PM Alerts: Delete: nothing to do [0.00/1.00GB, 162.6GB free] (840)
5/5/2024 12:14:04 PM New: Delete: nothing to do [5499/5500GB, 162.6GB free] (840)
5/5/2024 11:20:46 AM Stored: Delete: nothing to do [0/336 hrs, 0.00/1.00GB, 748.4GB free]
5/5/2024 7:24:43 AM Stored: Delete: nothing to do [0/336 hrs, 0.00/1.00GB, 748.6GB free]
5/5/2024 2:00:05 AM DB: Reindexing: 1706 records 11 msec
5/5/2024 2:00:05 AM DB: DBCompact: finished (2)
5/5/2024 Log: Retaining 13 files (39.6MB)
5/5/2024 App: Today's sunrise: 6:03 AM, sunset: 8:05 PM, timestamp=1714881600 (GMT+14400s)
5/4/2024 3:43:24 PM Stored: Delete: nothing to do [0/336 hrs, 0.00/1.00GB, 748.3GB free]

Any suggestions or can someone pass along to Ken?

[IAmATeaf]

You need to log a support ticket yourself, the more people that log an issue the more aware BI will be of how widespread the issue is.

[Skittles]

Support ticket won't help, Ken seems pretty willing to die on the hill that he's got it right and there is no problem, which I disagree with.

• My HW/OS has not changed, just BI version.

• Disk Queue Length, Throughput, Response Times at the OS level all look within spec for the drive and healthy.

• No tangible issues with playback, recording etc.

• BI benchmark shows 85MB/s max write speed, 3.2GB/s Read w/ a current rate of 5.51MB. I called out the fact that a 3.26GB/s read speed on a 256MB cache, 72000 RPM drive and a 10GB test file size was literally impossible, didn't seem to matter.

• I suggested implementing a control to alert the user of the "issue" once then give them some control to no longer receive the notification if they believe it's benign and everything is working correctly, didn't get traction.