386.9 upgraded and get immediate low NVRAM warning after.

[eastavin]

I just updated an RT-AC68u to 386.9 from 386.7_2 and am immediately getting a low NVRAM warning suggesting I factory reset. I have 6 reserved DHCP addresses, run FLEXQOS and an openvpn server that is currently inactive. Is this warning a bug, a new feature that did not exist or something else? I have not had any such warning the entire time I ran 386.7_2. I dont mind downgrading but thought I would ask if anyone else is seeing this before running off half-cocked.

Best regards,

Ed

 

[PeterR]

I just updated an RT-AC68u to 386.9 from 386.7_2 and am immediately getting a low NVRAM warning suggesting I factory reset. I have 6 reserved DHCP addresses, run FLEXQOS and an openvpn server that is currently inactive. Is this warning a bug, a new feature that did not exist or something else? I have not had any such warning the entire time I ran 386.7_2. I dont mind downgrading but thought I would ask if anyone else is seeing this before running off half-cocked.

Best regards,

Ed

Quite common for AC68U https://www.snbforums.com/threads/a...c2900-rt-ac68u-and-rt-ac86u.82209/post-808409

 

[bennor]

I just updated an RT-AC68u to 386.9 from 386.7_2 and am immediately getting a low NVRAM warning suggesting I factory reset.

A number of people have reported the issue in the alpha, beta and Final 386.9 and other threads. For example here, and here, and here. RMerlin explained the reason/cause for low NVRAM issue post 386.9 update previously in the Beta thread:

Low nvram notification is nothing new, and is becoming increasingly common for that old model as it's running out of nvram space with every new firmware adding new nvram settings. It's part of why Asus is leaving it on the 386 branch most likely, as 388 probably adds even more nvram settings.

This 9 years old model is starting to show its age, with Asus having added lots of new features to it since its initial launch in 2013.

I might have to start considering removing some things from the RT-AC68U to save some nvram space. Reducing the number of OpenVPN clients would be such a thing, as each client consumes around 1 KB of nvram even while not in use.

Do a hard factory reset, or try the some of the other suggestions for dealing with high NVRAM usage that were discussed in the following thread:
https://www.snbforums.com/threads/solved-high-nvram-usage.56848/#post-494736

Among them are moving one's manual DHCP reservations to /jffs/configs/dnsmasq.conf.add, or try using YazDHCP, and another is running the following to clean out unused NVRAM entries (as always use at your own risk).

for line in `nvram show | grep =$ `; do var=${line%*=}; nvram unset $var; done; nvram commit

Or

for line in `nvram show | grep ^[^=]*=$ `; do var=${line%*=}; nvram unset $var; done; nvram commit

 

[RMerlin]

I guess you could try unsetting every vpn_client5_* and vpn_client4_* settings to save some nvram if you don`t use them, however I'm not entirely sure if these wouldn't get recreated at next reboot - I never tested it. It would certainly cause issues if you were to ever try configuring either of these OpenVPN client instances.

 

[eastavin]

I tried out a number of these ideas. All had some effect on reducing nvram usage. Thank you all for the input. I was able to get rid of the warning on the ac68u simply by deleting 4 DHCP addresses. But it left me wondering.

The idea I liked the best in the end was just facing up and hard erasing nvram. This got the nvram usage down from a high 62 to 57 and change. After a further factory reset and reconfiguration of a openvpn server it settled at 58,678 bytes down a little over 4kbytes. So I am satisfied for now. I think I will just go and do this to the other units on this release now. Thanks again.