Asus rt86u "Format jffs at next boot"

[RMerlin]

Also to people from Poland flashing any firmware other than Asus-wrt stock will void the warranty on any asus device . Got this information from Asus themselfs. So would be nice to clear anything including jffs2 which is created exclusively on asus-merlin i believe or am i wrong because i dont seem to find info about this on stock/asus website.

The JFFS2 partition is present on stock firmware as well, it's where Asus stores various information like the AiProtection databases and such. On HND model part of the router settings are also saved there by Asus.

The warranty question will vary depending on who you ask. First-line tech support most likely don't know, so the answer will change every time you ask a different person. There has been cases where people asked Asus, and were told that it would not affect their warranty.

 

[podkaracz]

The JFFS2 partition is present on stock firmware as well, it's where Asus stores various information like the AiProtection databases and such. On HND model part of the router settings are also saved there by Asus.

The warranty question will vary depending on who you ask. First-line tech support most likely don't know, so the answer will change every time you ask a different person. There has been cases where people asked Asus, and were told that it would not affect their warranty.

Used the "technical question" function on asus website after registering my router . I have no idea who responded but it was stated as "asus poland stance" in that regard. Also even if stock has this partition it does not have possibility of erasing it , so here is my question how to do it without desoldering chip and flashing it with programer using files copied from working new model of in my case 86u.

 

[podkaracz]

What about this :

example: JFFS2_FEATURE_RWCOMPAT_DELETE
mkfs.jffs2 -rdir -oout.jffs2 -e128KiB
sumtool -iout.jffs2 -oout-sum.jffs2 -e128KiB

EBS should work with NOR and NAND, too. The rate of speedup is generally higher on NAND chips and on the chips where the erase block size is large.

During mount process if there is no summary node at the end of an erase block, the original scan process will be executed.

 

[ColinTaylor]

Also even if stock has this partition it does not have possibility of erasing it ...

That is what the "Initialize" option does at Administration > Restore/Save/Upload Setting > Factory default > Restore.

 

[podkaracz]

That is what the "Initialize" option does at Administration > Restore/Save/Upload Setting > Factory default > Restore.

So why is there additional option to erase jffs at next boot in merlin when initialize exist?

 

[ColinTaylor]

So why is there additional option to erase jffs at next boot in merlin when initialize exist?

Because Merlin added that option years ago at a time when the "Initialize" didn't exist in the stock firmware. It's just been left in since then.

 

[podkaracz]

Because Merlin added that option years ago at a time when the "Initialize" didn't exist in the stock firmware. It's just been left in since then.

I was thinking about it but leaving it without removing still suggest that its needed over initialize

 

[ColinTaylor]

I was thinking about it but leaving it without removing still suggest that its needed over initialize

We've already provided links to the relevant source code so that you can see for yourself if you don't believe us. Both options are doing the same thing. The problem (as with all filesystems) is that it can't format the device after it's been mounted. That is why it has to be done during the boot process before it's mounted. It does this by setting the nvram variable jffs2_format to 1.

 

[podkaracz]

We've already provided links to the relevant source code so that you can see for yourself if you don't believe us. Both options are doing the same thing. The problem (as with all filesystems) is that it can't format the device after it's been mounted. That is why it has to be done during the boot process before it's mounted. It does this by setting the nvram variable jffs2_format to 1. There is another variable which isn't exposed in any GUI setting called force_erase_jffs2. Setting that to 1 forces the router to format /jffs and follow it with a reboot.

I mean i know and i wrote it on purpose that my sentence will look like i dont believe you from logical point of view im just bumping this thread so maybe some1 will find proper commands for it . Thats the main reason i write here from time to time because as you can see im not able to find those commands for example for "force_erase_jffs2" which all that mess revolves around probably. So is "force_erase_jffs2" part of initialize process from web gui or its the main difference between merlin's "format at next boot" and initialize ? Again you write that it does the same thing and then you point a difference that one thing formats jffs before files are mounted suggesting that the second does it after files are mounted so one thing disclaims another when it comes to sentece that they both do the same. So the aim of both is to erase but one does it before file system is mounted and the other after so first one is better probably? Or i am interpreting your words in a wrong direction and format jffs at next boot = initialize but merlin forgot to remove format jffs at next boot?


#sorry if i keep this thread active longer than it should be and if my words are irrational and deny your arguments but this is my thought process and before i understand both options i might annoy you with constant asking even if what i write might seem dumb at times.

 

[ColinTaylor]

Sorry, I'm done with this conversation.

 

[RMerlin]

I kept the option there because not everyone wants to erase all their settings at the same time.

Sent from my ELE-L04 using Tapatalk

 

[podkaracz]

Ok thanks for your efforts i learnt more than i thought i would . Thread can be closed as everything seems to be clear right now . And sorry if my lack of knowledge is irritating sometimes.

 

[podkaracz]

I have my router working and fixed my connection issues but i like to close things up and i think ive found what i was looking for when i was creating that thread that is command line for some gui commands. So here are some of my findings. Remember dont use them if not needed since gui and router buttons are enough. I just wanted to learn something new as im newb and this forum has a lot of knowledge. I dont have problems with my router and those commands are for educational purposes because it took me some time to gather them and i was not able to find them directly in 1 thread.


Here is what ive found:

nvram get jffs2_format
nvram get jffs2_on
nvram commit
reboot

nvram unset jffs2_size
nvram set jffs2_on=1
nvram set jffs2_format=1
nvram commit
reboot


nvram set force_erase_jffs2=1 (seems like closest to what i wanted to achieve by creating this topic but you need to use wps reset to restore router functionality)
nvram commit
reboot


cd /jffs
rm -f on*.postconf
rm -rf /jffs/*
hnd-erase nvram
nvram commit
reboot

nvram set jffs2_clean_fs=1
nvram commit
reboot


And manual firmware flashing (not recommended)
hnd-write FIRMWARE.U linux && reboot


My lack of knowledge is heavy sometimes for some people but in this topic i see that @ColinTaylor is an expert as most of those commands comes from him. Also @L&LD has some nice nuclear reset guides.



Also i have 1 more question to @ColinTaylor and @L&LD . Is flashing firmware multiple times necessary ( like suggested in many older @L&LD guides ) or once is enough no matter how many resets we make? This is interesting as new firmwares will come out and im not sure how big of an impact dirty upgrading has on performance and if reflashing firmware actually solves any problems.

 

[L&LD]

On our own routers where we're there to monitor and fix things as needed, a Nuclear Reset, repetitive flashing and re-flashing of the same firmware, and other seemingly extreme suggestions are only required when the hardware or the network/router stops behaving as expected and we want it back to stable/reliable and fast.

I only wish I had that option for my customers. But for theirs and for my sake, I do the 'extreme' measures as a matter of course. If I have to go back to any of those customers, it is either a new device that needs to be integrated properly into the network or, the router hardware failed.

Not many routers fail, btw.