Anyone knows why this would keep spamming my syslog?:

Thu Jun  1 08:23:01 2017 kern.err kernel: [386214.432712] UBIFS error (ubi0:1 pid 25922): ubifs_readdir: cannot find next direntry, error -22
Thu Jun  1 08:23:01 2017 kern.err kernel: [386214.433019] UBIFS error (ubi0:1 pid 1595): make_reservation: cannot reserve 160 bytes in jhead 1, error -30
Thu Jun  1 08:23:01 2017 kern.err kernel: [386214.433024] UBIFS error (ubi0:1 pid 1595): ubifs_write_inode: can't write inode 334, error -30
Thu Jun  1 08:23:01 2017 kern.err kernel: [386214.460273] UBIFS error (ubi0:1 pid 25922): ubifs_read_node: bad node type (255 but expected 2)
Thu Jun  1 08:23:01 2017 kern.err kernel: [386214.469119] UBIFS error (ubi0:1 pid 25922): ubifs_read_node: bad node at LEB 12:12448, LEB mapping status 1
Thu Jun  1 08:23:01 2017 kern.err kernel: [386214.478997] Not a node, first 24 bytes:
Thu Jun  1 08:23:01 2017 kern.err kernel: [386214.482763] 00000000: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff                          ........................
Thu Jun  1 08:23:01 2017 kern.warn kernel: [386214.495778] CPU: 0 PID: 25922 Comm: luci Tainted: G        W       4.9.30 #0
Thu Jun  1 08:23:01 2017 kern.warn kernel: [386214.502941] Hardware name: Marvell Armada 380/385 (Device Tree)
Thu Jun  1 08:23:01 2017 kern.warn kernel: [386214.508978] [<c0016010>] (unwind_backtrace) from [<c0012220>] (show_stack+0x10/0x14)
Thu Jun  1 08:23:01 2017 kern.warn kernel: [386214.516843] [<c0012220>] (show_stack) from [<c020f9dc>] (dump_stack+0x7c/0x9c)
Thu Jun  1 08:23:01 2017 kern.warn kernel: [386214.524183] [<c020f9dc>] (dump_stack) from [<c01905fc>] (ubifs_read_node+0x228/0x258)
Thu Jun  1 08:23:01 2017 kern.warn kernel: [386214.532135] [<c01905fc>] (ubifs_read_node) from [<c01a8be8>] (ubifs_tnc_read_node+0x54/0xac)
Thu Jun  1 08:23:01 2017 kern.warn kernel: [386214.540696] [<c01a8be8>] (ubifs_tnc_read_node) from [<c01915c4>] (tnc_read_node_nm+0x11c/0x1fc)
Thu Jun  1 08:23:01 2017 kern.warn kernel: [386214.549519] [<c01915c4>] (tnc_read_node_nm) from [<c01944d8>] (ubifs_tnc_next_ent+0x164/0x1b0)
Thu Jun  1 08:23:01 2017 kern.warn kernel: [386214.558253] [<c01944d8>] (ubifs_tnc_next_ent) from [<c0188e18>] (ubifs_readdir+0x340/0x424)
Thu Jun  1 08:23:01 2017 kern.warn kernel: [386214.566728] [<c0188e18>] (ubifs_readdir) from [<c00df440>] (iterate_dir+0x78/0x124)
Thu Jun  1 08:23:01 2017 kern.warn kernel: [386214.574504] [<c00df440>] (iterate_dir) from [<c01b2494>] (ovl_dir_read_merged+0x94/0x1c8)
Thu Jun  1 08:23:01 2017 kern.warn kernel: [386214.582802] [<c01b2494>] (ovl_dir_read_merged) from [<c01b2958>] (ovl_iterate+0xc8/0x1e8)
Thu Jun  1 08:23:01 2017 kern.warn kernel: [386214.591100] [<c01b2958>] (ovl_iterate) from [<c00df440>] (iterate_dir+0x78/0x124)
Thu Jun  1 08:23:01 2017 kern.warn kernel: [386214.598700] [<c00df440>] (iterate_dir) from [<c00dfb48>] (SyS_getdents64+0x7c/0x11c)
Thu Jun  1 08:23:01 2017 kern.warn kernel: [386214.606562] [<c00dfb48>] (SyS_getdents64) from [<c000ed40>] (ret_fast_syscall+0x0/0x3c)
Thu Jun  1 08:23:01 2017 kern.err kernel: [386214.614700] UBIFS error (ubi0:1 pid 25922): ubifs_readdir: cannot find next direntry, error -22

EDIT: And I checked, disk space seems okay:

root@Scruffy:~# df -h
Filesystem                Size      Used Available Use% Mounted on
/dev/root                 7.5M      7.5M         0 100% /rom
tmpfs                   250.5M      1.8M    248.8M   1% /tmp
/dev/ubi0_1              52.6M      3.5M     46.4M   7% /overlay
overlayfs:/overlay       52.6M      3.5M     46.4M   7% /
ubi1:syscfg              70.2M    484.0K     66.1M   1% /tmp/syscfg
tmpfs                   512.0K         0    512.0K   0% /dev

(Last edited by farchord on 1 Jun 2017, 13:26)

Looks like sysupgrade is getting a bit of a face-lift. Maybe this will help making flashing a bit easier from LuCi?  I'm looking forward to finding out.

https://github.com/lede-project/source/ … 02217ab363

@farchord

I don't know a lot about those messages, but they don't look very good. Consider booting to the other partition, and flashing again.

@david, you must have reconfigured your web server again.  http is now working again, but not auto-forwarding to https.  Ideally the way it should work is if someone types in your web address without https://, it will auto-forward to https://.  The way you had it yesterday, if someone didn't type in https://, they got a 404.

I don't know how to fix it, but look at even how Luci-SSL/uhttpd works for maybe an example.  If you type in 192.168.1.1, it auto-forwards to https://192.168.1.1.

Edit: It just went dead after I posted this so I assume you are working on it

Edit2: It's now redirecting perfectly, great job!!

@Pasxalisk and others with issues, I don't mean to be an ass or come across as a jerk or anything, but before reporting an issue here, you should try the latest LEDE development branch build from https://downloads.lede-project.org/snap … u/generic/ and see if your issue exists there.  An issue should only be reported here if it doesn't exist in the official LEDE build, but does in @david's build.  Because if the issue exists in the official LEDE build., it's highly unlikely that @david can do anything about it.  And if you find an issue with the official LEDE build, it should be reported as an issue on their github or in a thread on the LEDE forums, so that it hopefully will be fixed.  Of course, if you find a work around for an issue, posting it here would of course be greatly appreciated.

Edit: I thought we had already established that your 1200V2 was defective due to its low speeds.  This signal level issue only seems to confirm that something is physically wrong with the router.

(Last edited by starcms on 2 Jun 2017, 00:52)

@starcms

I just started messing with redirection..  To get it working right I just had to use the rewrite engine method, and not the permanent redirect method.

@waleed.soliman

Is the re-connection of your wifi clients any fast now?

Glad someone is at least addressing this issue after so long.  Got my fingers crossed.  As I'm sure you know, one used to be able to flash any .img from LuCI, even the stock Linksys firmware.  As it is now, you can only flash .bin sysupgrades from LuCi.

Btw, for those who use dnscrypt, I submitted my first pull request (using github is a real bitch btw; for those that have never used it, you can only do the most very basic things from the web interface -- you need to download the git command line program to do practically everything), and version 1.9.5-4 is now out.  It updates the resolvers list (in the dnscrypt-proxy-resolvers package) which was over 6+ months out of date (although I would still recommend using a script like the one I posted a couple pages back to update it more frequently, because the list is updated on an almost daily basis) and adds an option for log_level for dnscrypt-proxy (in /etc/config/dnscrypt-proxy).  You can set it to 5, instead of the default of 6, to quiet much log spam.

1.9.5-4 is available now from the snapshot repo, but for those who use dnscrypt plugins, I would recommend waiting, because the version in the repo doesn't have plugin support enabled (unlike the versions that @david compiles).

(Last edited by starcms on 2 Jun 2017, 01:47)

@Pasxalisk

Yeah, I have the same issue with Luci reporting the wrong signal strength. I just ignore this since it doesn't really make or break the experience.

About the wireless scan? Yes, and it's annoying, it only fails to scan when you're connected to another network as a client, you would have to disconnect from the network temporarily to scan.

@davidc502,

i performed a full reset, then changed only the basic configurations (PPPOE, SSID, Encryption, Cipher, & DNS forwarding)

then upgraded again to r4222 and BOOM! everything is working fine!

i looked into the log, and Most of those entries of deauthentication are gone.

even the RADIUS entry seems better now:

Fri Jun  2 03:42:45 2017 daemon.info hostapd: wlan1: STA 70:81:eb:d1:f0:b8 RADIUS: starting accounting session F7606D2881075CF2
Fri Jun  2 03:42:45 2017 daemon.info hostapd: wlan1: STA 70:81:eb:d1:f0:b8 WPA: pairwise key handshake completed (RSN)
Fri Jun  2 03:42:46 2017 daemon.info dnsmasq-dhcp[4079]: DHCPREQUEST(br-lan) 192.168.1.175 70:81:eb:d1:f0:b8 
Fri Jun  2 03:42:46 2017 daemon.info dnsmasq-dhcp[4079]: DHCPACK(br-lan) 192.168.1.175 70:81:eb:d1:f0:b8 iPhone
Fri Jun  2 03:44:33 2017 daemon.notice hostapd: wlan1: AP-STA-DISCONNECTED c8:e0:eb:19:62:79
Fri Jun  2 03:44:33 2017 daemon.info hostapd: wlan1: STA c8:e0:eb:19:62:79 IEEE 802.11: disassociated
Fri Jun  2 03:44:33 2017 daemon.info hostapd: wlan0: STA c8:e0:eb:19:62:79 IEEE 802.11: authenticated
Fri Jun  2 03:44:33 2017 daemon.info hostapd: wlan0: STA c8:e0:eb:19:62:79 IEEE 802.11: associated (aid 1)
Fri Jun  2 03:44:33 2017 daemon.notice hostapd: wlan0: AP-STA-CONNECTED c8:e0:eb:19:62:79
Fri Jun  2 03:44:33 2017 daemon.info hostapd: wlan0: STA c8:e0:eb:19:62:79 RADIUS: starting accounting session E7D324EBEE5B8B08
Fri Jun  2 03:44:33 2017 daemon.info hostapd: wlan0: STA c8:e0:eb:19:62:79 WPA: pairwise key handshake completed (RSN)
Fri Jun  2 03:44:34 2017 daemon.info hostapd: wlan1: STA c8:e0:eb:19:62:79 IEEE 802.11: deauthenticated due to inactivity (timer DEAUTH/REMOVE)
Fri Jun  2 03:44:35 2017 daemon.info dnsmasq-dhcp[4079]: DHCPDISCOVER(br-lan) c8:e0:eb:19:62:79 
Fri Jun  2 03:44:35 2017 daemon.info dnsmasq-dhcp[4079]: DHCPOFFER(br-lan) 192.168.1.176 c8:e0:eb:19:62:79 
Fri Jun  2 03:44:36 2017 daemon.info dnsmasq-dhcp[4079]: DHCPREQUEST(br-lan) 192.168.1.176 c8:e0:eb:19:62:79 
Fri Jun  2 03:44:36 2017 daemon.info dnsmasq-dhcp[4079]: DHCPACK(br-lan) 192.168.1.176 c8:e0:eb:19:62:79 Waleed
Fri Jun  2 03:44:53 2017 daemon.info hostapd: wlan1: STA c8:e0:eb:19:62:79 IEEE 802.11: authenticated
Fri Jun  2 03:44:53 2017 daemon.info hostapd: wlan1: STA c8:e0:eb:19:62:79 IEEE 802.11: associated (aid 1)
Fri Jun  2 03:44:53 2017 daemon.notice hostapd: wlan1: AP-STA-CONNECTED c8:e0:eb:19:62:79
Fri Jun  2 03:44:53 2017 daemon.info hostapd: wlan1: STA c8:e0:eb:19:62:79 RADIUS: starting accounting session 28A2AFF374292A4A
Fri Jun  2 03:44:53 2017 daemon.info hostapd: wlan1: STA c8:e0:eb:19:62:79 WPA: pairwise key handshake completed (RSN)
Fri Jun  2 03:44:55 2017 daemon.info dnsmasq-dhcp[4079]: DHCPDISCOVER(br-lan) c8:e0:eb:19:62:79 
Fri Jun  2 03:44:55 2017 daemon.info dnsmasq-dhcp[4079]: DHCPOFFER(br-lan) 192.168.1.176 c8:e0:eb:19:62:79 
Fri Jun  2 03:44:56 2017 daemon.info dnsmasq-dhcp[4079]: DHCPREQUEST(br-lan) 192.168.1.176 c8:e0:eb:19:62:79 
Fri Jun  2 03:44:56 2017 daemon.info dnsmasq-dhcp[4079]: DHCPACK(br-lan) 192.168.1.176 c8:e0:eb:19:62:79 Waleed

Thank you David & Starcms , and keep the good work!

(Last edited by waleed.soliman on 2 Jun 2017, 05:01)

anyone?

I've definitely noticed more random dropouts on devices that never used to do it.

My Nexus 7 tablet can no longer see the 5GHz too. Tried starting from scratch on router + device but it still won't see it. Problem is it could be Lineage OS or LEDE and I don't have the time to work out which it is.

I've had issues with the latest driver on my v1 running the "other" OS that starts with DD-xxx

Poor 5GGz range and performance. Never saw it on older builds. I really think that all the tinkering they are doing with the driver to fix the 3200 may be messing with previously stable devices.

@david all right so I switched to Partition #2 (I was on 1), reflashed (Which I assume flashed to #1 again) and now it seems okay.

Odd. Not sure how this'd happen... I do make sure I always flash from a cabled connection to the router and not wifi...

Hi all,
I just bought a wrt3200acm (I had a still great wndr3700v2 but I wanted to permanently use my vpn account router side to cover the whole lan). I flashed latest davidc release and so far everything seems to work fine stability side.
My only complain so far is the wifi range, really poor (my older wndr3700v2 seemed even better). BTW it's not a big issue since I use powerline where I need coverage. I have some quetions.
1) How can I flash an update without lose the original stock firmware on the other partition AND my configuration? If I got it right if I switch to the original partition and flash it erase configuration. Is that right?
2) I successfully configured my vpn on router so my whole lan traffic passes thru VPN server but when I acquire public VPN ip I lose my original ip functionality. This means I can't reach my router remotely anymore since my dyndns points to my original ip and it's not reachable. Is that normal? I mean when I use VPN public IP my original ip stops working? I've never experienced anything like that and since my lack of experience I don't know if it's normal.
3) I tried to configure a vpn client on my cellphone. It works fine when I use it through wifi connection but it doesn't when it's in LTE. is there a way to make it work? If needed I can proivide logs and configurations (now I'm at work).
4) Is there a way to selectively route services based on dns? My netflix account and Sky Go don't work with my VPN ip.

Thanks for any advice and to davidc and others for your efforts.

(Last edited by puppinoo on 2 Jun 2017, 13:43)

1) The 3200ACM swaps firmware everytime you flash on it. So your last firmware is always intact. If you don't want to wipe a partition, to my knowledge, before you reflash, you need to swap to the other partition beforehand. To clarify: If you are let's say on Partition 1 and you do a firmware update. It will flash onto Partition 2 and reboot onto that new partition. And so on.

I know but I don't want to lose my configuration every time so I was wondering if flashing from the other partition erases it.

Thanks.

If you have flashed only once.

SSH into your router and issue the commands below, the router will boot into the first partition, which will be Linksys, if you have flashed only once. You will lose your LEDE config when you flash again from the Linksys GUI. Personally, I prefer to flash a new image and configure form scratch. Every time.

fw_setenv boot_part 1
reboot

I always keep the Linksys firmware in a partition. These routers are unbrickable if you do that.

(Last edited by mojolacerator on 2 Jun 2017, 14:30)

Thanks. Can I at least save my configuration zip and resrtore later or it's discouraged?

maybe using sysupgrade can upgrade the actual partition keeping the settings intact?
I did it with older Openwrt versions.

Regards.

This will write over the partition. You will then have your upgraded LEDE image in partition one. And the older LEDE image in partition two.

No I believe that sysupgrade will also upgrade to the 2nd partition. And if you install from stock linksys firmware I think there's no guarantee.

But yeah if you backup your config before flashing you should be fine. You'll be safe whatever happens.

So to be sure If I am on stock firmware partition, using sysupgrade will still flash to the OTHER partition?
If so maybe I still can use the option to flash fw keeping the settings (?).
Thanks a lot for info.
Il everything goes south Ill try using the backup.

If you are on Linksys firmware and you flash LEDE, you will have LEDE in partition 2 and Linksys in partition 1.
If you sysupgrade LEDE in partition 2, you will overwrite Linksys in partition 1 and have LEDE in both.

Nice, Thanks. I hope to solve my issues related to VPn and then it's all fine. I also wonder if the weak wifi power is related to driver still problematic so hopefully will be solved or it's hardware related. But overall it's ok.