Slowness, No Network access then ZRAM errors in dmesg

Hi issues beginning to crop up across multiple xaviers.
Issue will manifest itself as the ssh connection will drop out, no IO activity i.e. cameras connected will stop and restapi server will become unreachable and then after 20-30 seconds return to normal.
dmesg logs are then full of the following messages:
[ 3550.103183] Write-error on swap-device (252:7:3515888)
[ 3550.103191] zram: Error allocating memory for compressed page: 439516, size=4096
[ 3550.103193] zram: Error allocating memory for compressed page: 75628, size=4096
[ 3550.103197] zram: Error allocating memory for compressed page: 439516, size=4096

I don’t know what to make of this or where to start investigating, any help? Im on Jetpack 4.2.3 rev1 and do not wish to upgrade unless this issue has been solved in a later jetpack version.

This issue tends to happen (but may be unrelated) around time periods where large files (1.1gb) are transferred off the NVME drive to an External drive plugged into the eSATA. It happens infrequently but reliably as in once a day. There is also alot of network traffic on these units, a steady stream of high res images being sent across the network, so in terms of bus load theres a bit.
But i’ve only just started recently seeing this with no major changes, but the same image has been copied and flashed to other xaviers with the same symptoms and dmesgs.

Sorry for the late response, so could you try with other devices to see if issue still presents?

There is no update from you for a period, assuming this is not an issue any more.
Hence we are closing this topic. If need further support, please open a new one.


Could you give a detail information of what are needed to reproduce this issue?