[WENED]

Before anyone jumps down my throat, yes I have raised this issue before and mentioned it on a number of occasions since.

I have tried numerous new builds on my remote 1900V1 and have always had to revert to r31924 as the connection will drop and periodically reboot. On the last attempt r44187 the router rebooted after only a half hour running and later when lost the remote connection I could not access it through the 5Ghz WDS. The logs indicated a continuous retry of connection but was unsuccessful. I was able to gain access on the 2.4Ghz and switch partitions back to r31924. Absolutely no problems in the past 36 hours.
Upon checking my other remote WDS 1900V1 I noticed it had rebooted about 16 hours ago. There was no log entry to determine the cause, just a reconnection normal upon reboot. This router is on a UPS so power is not in question. It however is at the marginal distance with a signal average -72db. I have left it at r44187 as I only have two cameras on lan with it, however if I continue to have problems I will revert it also.
Some time ago I suspected that the 2.4Ghz band operating on the remotes was causing the problem, but now with losing the second one which doesn't even have 2.4Ghz (another story) I have dissolved that theory.
My third 1900V1 is still at r44187 and working flawlessly as WDS AP with a half dozen 2.4Ghz mixed clients including ESP8266 devices.
I am not overly concerned for my applications as the using the old r31924 works fine but;
although I cannot test other wrt routers except for my 3200 which is working fine but in relatively close proximity, I am re-posting this in the event that other users are experiencing strange behavior on remote routers in Client or WDS configurations.
**** UPDATE ****
Installed r44191 on my 1200V1 and replaced the one 1900V1 to see if problem occurs on it. Signal is -80db so it will be interesting.

As best I can determine there is problem with running remote connections where the signal strength is less than -70db and a subroutine within the code of newer builds (past couple of years as r31924 works) is causing the remote to do a re-connection and if after a number of re-tries do a full reboot. This is the only conclusion I have after fighting with this for the past couple of years and doing everything imaginable to stop the reboots (better power supply, UPS, etc.) without success until I installed r31924.
So, if you are having random reboots/Reconnects on a remote consider finding a an older build. r31924 I believe was only available for 1900V1 but maybe for other routers but that is a question that needs to be directed to the Gurus. Better yet for someone who is code savvy taking a look at r31924 and comparing to all later build might yield the cause of this problem.

[WENED]

12 hours later and the 1200 is still connected and no problems encountered.
One other observation I have not mentioned;
The 1900 V1 is distinctive whereas it uses a different CPU than the 1200 & later versions 1900's, also it has only half the RAM that the other routers have.

Also I keep mentioning r31924 but I know it is not in the Router Database or Beta lists. Here is the link to the build if anyone is interested (many thanks to Spuriousoffspring for making it available), I would still chasing ghosts if it were not for this build.

https://forum.dd-wrt.com/phpBB2/viewtopic.php?t=310706

[WENED]

Well as usual I can't leave things alone.

Upgraded all routers to r44191 with IE11 reset to default and re-configured as required.
rebooted entire system (all routers).
Been running only a couple of hours but so far no hangs/reboots/re-connects on any devices.
2.4Ghz performing exceptionally well where all devices are connecting N mode except my old printer (Legacy).

Will be monitoring for now. Hoping I can keep this configuration and not revert my 1900's back to r31924 but I've been here before so I am pessimistic.

***** UPDATE ******

A little over two hours in and a reboot occurred.
Noticed I had not set Local DNS and Gateway adresses so I entered the Master Address and applied. Was unable to connect to router afterwards. Power booted router and all restarted again.
May not have been recording Logs to Papertrail as Gateway and Local DNS were not set. Will monitor now with all redone. Maybe will see a sequence when the reboot happens.

[WENED]

Not quite 12 minutes in and router disappeared. Reboot got router going again but was unable to connect to WDS AP showing the following errors in log;

Aug 16 16:47:55 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be MLME: auth request, signal -71 (Accepted)
Aug 16 16:47:55 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be IEEE 802.11: authenticated
Aug 16 16:47:55 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be MLME: assoc request, signal -70 (Accepted)
Aug 16 16:47:55 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be IEEE 802.1204.1: associated (aid 1)
Aug 16 16:48:05 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be IEEE 802.11: deauthenticated due to local deauth request
Aug 16 16:48:05 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be MLME: auth request, signal -71 (Accepted)
Aug 16 16:48:05 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be IEEE 802.11: authenticated
Aug 16 16:48:05 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be MLME: assoc request, signal -72 (Accepted)
Aug 16 16:48:05 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be IEEE 802.11: associated (aid 1)
Aug 16 16:48:14 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be MLME: auth request, signal -71 (Accepted)
Aug 16 16:48:14 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be IEEE 802.11: authenticated
Aug 16 16:48:14 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be IEEE 802.11: deauthenticated due to local deauth request
Aug 16 16:48:14 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be IEEE 802.11: Station tried to associate before authentication (aid=-1 flags=0x0)
Aug 16 16:48:19 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be MLME: auth request, signal -70 (Accepted)
Aug 16 16:48:19 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be IEEE 802.11: authenticated
Aug 16 16:48:19 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be MLME: assoc request, signal -70 (Accepted)
Aug 16 16:48:19 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be IEEE 802.11: associated (aid 1)
Aug 16 16:48:28 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be IEEE 802.11: deauthenticated due to local deauth request
Aug 16 16:48:37 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be MLME: auth request, signal -72 (Accepted)
Aug 16 16:48:37 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be IEEE 802.11: authenticated
Aug 16 16:48:37 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be MLME: assoc request, signal -71 (Accepted)
Aug 16 16:48:37 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be IEEE 802.11: associated (aid 1)
Aug 16 16:48:46 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be IEEE 802.11: deauthenticated due to local deauth request
Aug 16 16:49:07 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be MLME: auth request, signal -71 (Accepted)
Aug 16 16:49:07 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be IEEE 802.11: authenticated
Aug 16 16:49:07 XXX.XXX.XXX.XX hostapd ath1: STA c0:56:27:68:4b:be MLME: assoc request, signal -70 (Accepted)

Rebooted the WDS AP and the Station again. Operation started successfully. Will continue to monitor.

[WENED]

What a mess!

Router ran for about 8 hours then rebooted. Reboots occurred a few more times at varying times 30 min to hours. Eventually by morning when I checked the status the WDS link was in a loop trying to connect.
Since I had all routers recording logs to Papertrail I turned off all except the failing unit.
Rebooted the 1900V1 and will be monitoring. I will provide a copy of the logs later, possibly hours or a day but will have a history of the fail.
Once done, if I cannot see any specific resolution, as I have tried most everything, I will revert the router back to r31924.

[WENED]

I apologize for being annoying about this specific issue. I have been playing around with different settings and configurations over the past couple of days and may finally have a clue what is happening, although no idea how to fix it.
Yesterday I upgraded all routers to r44213 and changed the two remote 1900's from WDS Station to Client Bridged. This created some minor problems but easily fixed with configuration of my devices.
What I have noticed from the logs is a reboot occurs consistently whenever the time server is accessed (every hour) and the time appears to be different from that within the router. Sometimes this only happens after 8+ hours, other times it happens as much as every hour. I have no idea what is causing this time difference or why the router thinks it has to reboot rather than simply update the time.

Aug 20 17:08:13 XXX.XXX.XXX.XXX kernel [ 1411.936277] ieee80211 phy0: Mac80211 start BA ec:3d:fd:e1:a1:5e
Aug 20 17:16:16 XXX.XXX.XXX.XXX ntpclient Time set from 2.pool.ntp.org [198.100.155.119].
Aug 20 17:16:16 XXX.XXX.XXX.XXX process_monitor cyclic NTP Update success (servers 2.pool.ntp.org 212.18.3.19 88.99.174.22)
Aug 20 17:16:16 XXX.XXX.XXX.XXX process_monitor Restarting cron (time sync change)
Aug 20 17:16:16 XXX.XXX.XXX.XXX logger : cron : daemon successfully stopped
Aug 20 17:16:16 XXX.XXX.XXX.XXX logger : cron : daemon successfully started
Aug 20 17:16:16 XXX.XXX.XXX.XXX cron (CRON) STARTUP (fork ok)
Aug 20 17:16:16 XXX.XXX.XXX.XXX logger : process_monitor : daemon successfully stopped
Aug 20 17:16:16 XXX.XXX.XXX.XXX process_monitor process_monitor : cleanup timers
Aug 20 17:16:16 XXX.XXX.XXX.XXX logger : process_monitor : successfully started
Aug 20 17:16:16 XXX.XXX.XXX.XXX process_monitor We need to re-update after 3600 seconds
Aug 20 17:16:16 XXX.XXX.XXX.XXX process_monitor process_monitor : set timer: 3600 seconds, callback: ntp_main()

As can be seen by the previous upon the time check a difference of 8 minutes occured followed by cron restarting and actually rebooting. This seems to be the time difference in most reboots 7 - 8 minutes.

Other than this observation I have not been able to determine the cause of the reboots. I hope that maybe someone out there has a clue what is happening although I can live with the problem it is somewhat annoying and may be causing other problems not apparent to me.

[vt]

i'm having a very similar issue with my pair of WRT 3200's.. so eerily familiar that I had to register an account to reply to this.

I hadn't thought to check the time differences, but i'm not streaming syslog either, that's a good idea.

I've only noticed it reboots (and not just drop connection randomly) since reading this thread.

Firmware: DD-WRT v3.0-r44340 std (09/10/20)

The discoonects are happening much more frequently for me though, it will stop pinging every few minutes.. but only on the WDS station. the other one seems fine. =(

[WENED]

[pbphoto]

Agreed - 44048 is where I'm at on my 3200ACM and 1900ACSv1. I thought 44406 was great but after a few hours, wifi devices started hanging up.