What's new

Scribe Asuswrt-Merlin 388.2 is now available for select models

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

Weblee2407

Regular Contributor
i get issues when it’s that high so I look to see what’s going on. The web gui is very sluggish, video starts to go out of focus and resharpen, Alexa has to think about it. I haven’t been able to get long term stability Especially with dnsmasq and skynet hanging and crashing.

I suspect AMTM entware packages are not stable so this weekend wipe and manual config with SSD reformat as well. i have not seen any recommendations on under which circumstances an entware wipe is recommended.
Well I finally found (and changed) something that had a positive impact on my AX6000's performance, consistency and memory usage.

I did a wipe/reinstall from scratch of Entware packages. No change at first - memory usage went up pretty quickly and was in the 90's within a couple of hours. I uninstalled components and after each one was reinstalled I checked the logs for dnsmasq events and webpage responsiveness. When I installed uiScribe I noticed a message I probably had overlooked because of the color scheme I had been using in my SSH client.

WARNING: Your configuration file uses an obsoleted keyword, please update your configuration;
keyword='stats_freq', change='Use the stats() block. E.g. stats(freq(1));', location='/opt/etc/syslog-ng.conf:19:5
'

I updated the syslog-ng.conf line 19 from something like stats_freq(21600) to stats(freq(21600)) and uninstalled / reinstalled uiScribe.

Went into the scribe menu and performed a rl - reload and a rs - restart syslog-ng and it was immediately clear this had been an issue. The webpage is again screaming fast, memory usage started at 56 and is just at 64 5 hours later! All the logs are being properly filtered as well. No dnsmasq tainted messages or skynet crashes.

I would like to ask the professional programmers of this forum to explain, if they can, how this one setting was impacting so much of my system.
I will admit some broken things were probably fixed by reinstalling the packages I use, but the primary issues were not resolved until I made this final edit and then reinstalled uiScribe.
 
Well I finally found (and changed) something that had a positive impact on my AX6000's performance, consistency and memory usage.

I did a wipe/reinstall from scratch of Entware packages. No change at first - memory usage went up pretty quickly and was in the 90's within a couple of hours. I uninstalled components and after each one was reinstalled I checked the logs for dnsmasq events and webpage responsiveness. When I installed uiScribe I noticed a message I probably had overlooked because of the color scheme I had been using in my SSH client.

WARNING: Your configuration file uses an obsoleted keyword, please update your configuration;
keyword='stats_freq', change='Use the stats() block. E.g. stats(freq(1));', location='/opt/etc/syslog-ng.conf:19:5
'

I updated the syslog-ng.conf line 19 from something like stats_freq(21600) to stats(freq(21600)) and uninstalled / reinstalled uiScribe.

Went into the scribe menu and performed a rl - reload and a rs - restart syslog-ng and it was immediately clear this had been an issue. The webpage is again screaming fast, memory usage started at 56 and is just at 64 5 hours later! All the logs are being properly filtered as well. No dnsmasq tainted messages or skynet crashes.

I would like to ask the professional programmers of this forum to explain, if they can, how this one setting was impacting so much of my system.
I will admit some broken things were probably fixed by reinstalling the packages I use, but the primary issues were not resolved until I made this final edit and then reinstalled uiScribe.
MANY thanks for your contribution - I too was doing battle with Scribe and had failed to notice the error picked up by Jack's add on uiScribe script.
My issue showed up mainly within the Skynet log which was misbehaving and not showing up as it should in the webui.
 
Well, this sent me down a rabbit hole.

1. I haven't upgraded the entware package. The 4.x versions are supposed to be backward compatible with the 3.x versions, except for the python options. So this is supposed to be just a warning.
2. The error message is a syslog-ng error message.
3. There is no documentation yet, so far as I can see, for the 4.x package. One identity is still at 3.38, and I wonder if they are continuing on. syslog-ng now looks like it is being maintained by Axoflow, so I'm really confused about who is on first.
4. And I just now noticed that all the documentation relates to global obtions in the form of, e.g., stats-freq, not stats_freq.


EDIT: I upgraded to syslog-ng to 4.2 and got the warning, but otherwise fine. I edited the .conf file to replace all the "_" with "-", and that was fine too. I changed the stats-freq to the stats() format, and that worked fine. I haven't noticed any differences in speed or memory use.
 
Last edited:
Well I finally found (and changed) something that had a positive impact on my AX6000's performance, consistency and memory usage.

I did a wipe/reinstall from scratch of Entware packages. No change at first - memory usage went up pretty quickly and was in the 90's within a couple of hours. I uninstalled components and after each one was reinstalled I checked the logs for dnsmasq events and webpage responsiveness. When I installed uiScribe I noticed a message I probably had overlooked because of the color scheme I had been using in my SSH client.

WARNING: Your configuration file uses an obsoleted keyword, please update your configuration;
keyword='stats_freq', change='Use the stats() block. E.g. stats(freq(1));', location='/opt/etc/syslog-ng.conf:19:5
'

I updated the syslog-ng.conf line 19 from something like stats_freq(21600) to stats(freq(21600)) and uninstalled / reinstalled uiScribe.

Went into the scribe menu and performed a rl - reload and a rs - restart syslog-ng and it was immediately clear this had been an issue. The webpage is again screaming fast, memory usage started at 56 and is just at 64 5 hours later! All the logs are being properly filtered as well. No dnsmasq tainted messages or skynet crashes.

I would like to ask the professional programmers of this forum to explain, if they can, how this one setting was impacting so much of my system.
I will admit some broken things were probably fixed by reinstalling the packages I use, but the primary issues were not resolved until I made this final edit and then reinstalled uiScribe.
@elorimer is correct, despite the warning message, I couldn't find any actual documentation on the change. Syslog-ng should have refused to start if it was no longer a valid entry, so I'm left to assume it's been deprecated, but not invalidated. As to why that would cause massive slowdown, I can only guess syslog-ng was massively spamming the syslog for some reason. I did post a sed command to fix the configuration file a couple days ago in the scribe 3.x_y thread: https://www.snbforums.com/threads/scribe-3-x_y-syslog-ng-and-logrotate-installer.84720/post-845976, but I wasn't experiencing any massive slowdown or anything like that on my router before I fixed it. Yours is the first report I've seen of massive slowdown.
 
OK..I have everything logging properly now. It seems my AMTM install muscle memory was creating a swap file BEFORE I installed any add-ons, AND also it seems SkyNet either does not see or will not use an existing swap file when it is being installed. Memory usage is just over 90, but the GUI is still responsive and the log filtering etc. are working.
 
So this was a swap issue, not a syslog-ng or scribe issue.
 
I've never seen any issues with creating a swap file with amtm (and rebooting the router afterward). Hard to believe this would be an issue.
 
I always reboot when prompted. Maybe it's my storage device?
 
Possibly. What is it specifically?
 
Western Digital Elements SE SSD USB3

Update: A friend who also has one tells me it will go into power save after 10 minutes of inactivity. Well first I have heard of that so starting to see if there is any confirmation. I don't think, with a swap file on it, there is a 10-minute period of no !/O.
I can't find anything to support that statement.
 
Last edited:

Similar threads

Latest threads

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top