[384.9_Alpha - builds] Testing all variants.

[huzaimie360]

just update and do clean install firmware and also freshJR QOS script default configuration...
my wan is PPPo


so far playing the division PVP in the darkzone killing agent, my bullet hit registration are on the spot no lag at all so far..

 

[RMerlin]

Running alpha 2 on my RT-AC86U, just to make sure you are aware the dcd crashes are back.

I had temporarily downgraded the Trend Micro components to the previous version to fix it, however that caused other issues, so the RT-AC86U is back to the 45149 component. And tests at downgrading only the dcd exe itself failed.

So, people will have to live with it until Asus fixes it - same as people on stock firmware.

 

[RMerlin]

I implemented something similar in sh for Skynet

Had fun working with the IPv6 support?

I spent the whole day working on it yesterday. Trend Micro's proc interface truncates the last two bytes of the IPv6 (in addition to formatting it in a very weird way). So I have to fake the last two bytes of the address for now...

 

[Adamm]

Had fun working with the IPv6 support?

I spent the whole day working on it yesterday. Trend Micro's proc interface truncates the last two bytes of the IPv6 (in addition to formatting it in a very weird way). So I have to fake the last two bytes of the address for now...
View attachment 15884

I noticed some funky stuff going on there... I went for the lazy approach for the time being with a generic “IPv6” placeholder

80+ characters for the combined IP fields doesn’t fit so great in an SSH terminal with all the other data which is also a PITA.

 

[RMerlin]

80+ characters for the combined IP fields doesn’t fit so great in an SSH terminal with all the other data which is also a PITA.

Yeah, formatting is a pain. I barely managed to squeeze it on my display (and I had to implement IPv6 address compression since the proc interface gives an expanded address).

You could always do it in two separate passes: parse the ipv4 entries, then the ipv6 entries. You can then use a different layout for IPv6.

I forwarded the info about the truncated bytes. Haven't heard any response yet, with any luck it might reach Trend Micro, and they might consider it important enough for them to fix... Otherwise, I might have to try to correlate the data with what's in Netfilter, and extrapolate the missing bytes from there if possible.

 

[huzaimie360]

THANK YOU MERLIN... THANK YOU FRESHJR

 

[RMerlin]

Thoughts?

 

[dave14305]

I like the previous text in the colored box for quicker identification/association with the QoS categories. The left-justified color box is too far from the text (at least I personally need to scan left-to-right/right-to-left to see them both).

 

[RMerlin]

I like the previous text in the colored box for quicker identification/association with the QoS categories. The left-justified color box is too far from the text (at least I personally need to scan left-to-right/right-to-left to see them both).

I find good points and bad points in both designs (and due to the way the table is rendered, I have very limited control over its content).

The square version is easier to quickly identify, but makes it impossible to sort by application name (because I have to put the CSS before the actual text, that means the sort will be done based on the CSS text rather than the Application text).I figured out a way to sort it by either application name or categories

The round version looks more polished, it allows me to sort by Application name (since the CSS is inserted after the text, and moved to the left through a float attribute), but it's less visual, and will generate word wraps with long application names (due to the larger width of the data).
Ideally, I would have wanted the entire table cell to be filled with color (like the pie chart legends), but that's not an option due to the table rendering code (or would require a lot of custom code, which I'd rather not waste time on).

 

[dave14305]

The square version is easier to quickly identify, but makes it impossible to sort by application name (because I have to put the CSS before the actual text, that means the sort will be done based on the CSS text rather than the Application text).

But does that mean we could sort by color so that all like applications within a category are together?

 

[RMerlin]

But does that mean we could sort by color so that all like applications within a category are together?

Yes, but in no "real" order, since it's sorted by RGB value.

I have however just changed that, replacing styling with CSS categories, so now they are in the same order as the categories in the charts, since the sort is done on a string that begins like this:

<span class="cat0"...
<span class="cat1"...
<span class="cat2"...
etc...

I also figured out a way to trick it to sort by actual app description by inserting it in a "title" attribute before the class.

So beside figuring out between the coloured dot and the large uneven square rectangle, I can now chose between sorting by Application or by category. Between sorting by categories and sorting by Application name, I favor sorting by name since this is the visual data. Sorting by categories would be unintuitive to end-users, as it would visually be more a grouping option than an alphabetical sort.

 

[RMerlin]

Just had a chat with a friend who does web application development for a living (among other things). He made various suggestions which IMHO fixed what was wrong with the original square label version.

 

[shark]

I don't know if this has been reported for the lasted alpha for the RT-AC86U, but here we go:

Jan 14 19:09:49 WLCEVENTD: eth5: Disassoc 00:0A:F5:FD:05:44
Jan 14 19:09:53 WLCEVENTD: eth5: Assoc 00:0A:F5:FD:05:44
Jan 14 19:09:53 kernel: [BLOCKED - INBOUND] IN=ppp0 OUT= MAC= SRC=176.119.4.35 DST= LEN=40 TOS=0x00 PREC=0x00 TTL=240 ID=38235 PROTO=TCP SPT=57328 DPT=9283 SEQ=518387245 ACK=0 WINDOW=1024 RES=0x00 SYN URGP=0 MARK=0x8000000
Jan 14 19:10:12 kernel: dcd[21178]: unhandled level 3 translation fault (11) at 0x00000000, esr 0x92000007
Jan 14 19:10:12 kernel: pgd = ffffffc0017a8000
Jan 14 19:10:12 kernel: [00000000] *pgd=000000001117c003, *pud=000000001117c003, *pmd=00000000150d6003, *pte=0000000000000000
Jan 14 19:10:12 kernel: CPU: 1 PID: 21178 Comm: dcd Tainted: P           O    4.1.27 #2
Jan 14 19:10:12 kernel: Hardware name: Broadcom-v8A (DT)
Jan 14 19:10:12 kernel: task: ffffffc014796ac0 ti: ffffffc01e95c000 task.ti: ffffffc01e95c000
Jan 14 19:10:12 kernel: PC is at 0xf7356f44
Jan 14 19:10:12 kernel: LR is at 0x1dc74
Jan 14 19:10:12 kernel: pc : [<00000000f7356f44>] lr : [<000000000001dc74>] pstate: 600e0010
Jan 14 19:10:12 kernel: sp : 00000000ffd7cb48
Jan 14 19:10:12 kernel: x12: 000000000009ff10
Jan 14 19:10:12 kernel: x11: 00000000f65ff024 x10: 00000000000a02b4
Jan 14 19:10:12 kernel: x9 : 00000000f65ff898 x8 : 00000000000a076c
Jan 14 19:10:12 kernel: x7 : 00000000f65ff8d0 x6 : 00000000000a0766
Jan 14 19:10:12 kernel: x5 : 0000000000000000 x4 : 00000000f65ff87c
Jan 14 19:10:12 kernel: x3 : 0000000000000000 x2 : 00000000ffd7cb24
Jan 14 19:10:12 kernel: x1 : 000000000007c674 x0 : 0000000000000000

Jan 14 19:10:52 kernel: dcd[21867]: unhandled level 3 translation fault (11) at 0x00000000, esr 0x92000007
Jan 14 19:10:52 kernel: pgd = ffffffc011cfb000
Jan 14 19:10:52 kernel: [00000000] *pgd=0000000011379003, *pud=0000000011379003, *pmd=0000000014661003, *pte=0000000000000000
Jan 14 19:10:52 kernel: CPU: 1 PID: 21867 Comm: dcd Tainted: P           O    4.1.27 #2
Jan 14 19:10:52 kernel: Hardware name: Broadcom-v8A (DT)
Jan 14 19:10:52 kernel: task: ffffffc011177480 ti: ffffffc0157e4000 task.ti: ffffffc0157e4000
Jan 14 19:10:52 kernel: PC is at 0xf74f5f44
Jan 14 19:10:52 kernel: LR is at 0x1dc74
Jan 14 19:10:52 kernel: pc : [<00000000f74f5f44>] lr : [<000000000001dc74>] pstate: 600e0010
Jan 14 19:10:52 kernel: sp : 00000000ffe21fc8
Jan 14 19:10:52 kernel: x12: 000000000009ff10
Jan 14 19:10:52 kernel: x11: 00000000f697c024 x10: 00000000000a02b4
Jan 14 19:10:52 kernel: x9 : 00000000f697c898 x8 : 00000000000a076c
Jan 14 19:10:52 kernel: x7 : 00000000f697c8d0 x6 : 00000000000a0766
Jan 14 19:10:52 kernel: x5 : 0000000000000000 x4 : 00000000f697c87c
Jan 14 19:10:52 kernel: x3 : 0000000000000000 x2 : 00000000ffe21fa4
Jan 14 19:10:52 kernel: x1 : 000000000007c674 x0 : 0000000000000000

 

[RMerlin]

Jan 14 19:10:12 kernel: CPU: 1 PID: 21178 Comm: dcd Tainted: P O 4.1.27 #2

Yes, extensively discussed, was mentioned just a few posts above.

 

[shark]

Yes, extensively discussed, was mentioned just a few posts above.

Thx and sry, for not reading the other posts, i don`t have the time right now to do that, i need to leave! So, im really sry for posting this issue again.

 

[skeal]

Has anyone else had trouble with the client list on the network map page not showing all connected devices? It will refresh and then still not show all devices. Right now mine shows 2 devices, I have at least 14. Now I know @RMerlin has no access to that code but, does anyone else have this issue before I reset?

 

[Tekneek]

Has anyone else had trouble with the client list on the network map page not showing all connected devices? It will refresh and then still not show all devices. Right now mine shows 2 devices, I have at least 14. Now I know @RMerlin has no access to that code but, does anyone else have this issue before I reset?

That is happening for me. The count is also wrong. It stops at 19, yet there are over 30 devices connected and functioning properly. Since it appears to be only a GUI problem, I've not worried about it.

I have been running 384.9_alpha2-g2ee0afa1c on my RT-AC3200 for about 8.5 hours.

 

[D_Day]

Has anyone else had trouble with the client list on the network map page not showing all connected devices? It will refresh and then still not show all devices. Right now mine shows 2 devices, I have at least 14. Now I know @RMerlin has no access to that code but, does anyone else have this issue before I reset?

Same thing with my Ac86u.
Everything is working fine though.

 

[skeal]

Has anyone else had trouble with the client list on the network map page not showing all connected devices? It will refresh and then still not show all devices. Right now mine shows 2 devices, I have at least 14. Now I know @RMerlin has no access to that code but, does anyone else have this issue before I reset?

That is happening for me. The count is also wrong. It stops at 19, yet there are over 30 devices connected and functioning properly. Since it appears to be only a GUI problem, I've not worried about it.

I have been running 384.9_alpha2-g2ee0afa1c on my RT-AC3200 for about 8.5 hours.

Anyone else?

 

[Sonyrolfy]

Anyone else?

Had it too, Clicked on Interface, wait 3 sec and they all show up.