Yes, only temporary downtime.
my device is running fine now.
i got the ssh bin file on my android phone. on normal pc I always get above error!
the xiaomi mi wifi 3g router works great so far!
128MB FLASH! so great, you can install wathever you need/like and still 87% freespace 
i tried vlans today, it did not work for me. has anyone else tried vlans on this device?
Did you fill up all vlans from 1 up? They don´t have to be connected (CPU tagged not needed) but they have to exist.
deleted is not working
Just installed latest LEDE build (snapshot r5369-2dc4852) in my brand new Xiaomi 3G router as I wanted to use it as Wifi repeater and being able to set a different IP range for the Xiaomi and unfortunatelly this can't be done with the default firmware.
I have the setup working but Wifi performance for devices connected to the repeated Wifi radio created by the Xiaomi router is very poor. The upload speed is great but the download speed is very low. Connecting to the main router gives me about 50Mb/second but doing the same test with the same device connected to the Xiaomi gives no more of 3 or 4Mb 
Also for some reason all the devices connected to the Xiaomi wifi lost internet access after a few minutes, wifi is not dropped just no access to internet, no problem when connected to the main router. To access to internet again I must to disconnect from the Xiaomi and connect again, then I have internet again for some minutes or even less than 30 seconds.
If someone can give me some pointers on how to solve this I would thank you a lot.
Update The snapshots are working. For some unknown reason sysupgrading to a newer build was making the router unstable. But after a factory reset the build became fairly stable. Read Xiaomi WiFi Router 3G - #222 by hammer for details.
Original post below:
Just a little warning as there might me something wrong with the lastest builds for the Xiaomi 3G.
My router was running perfectly for the last 10+ days and tonight i decided to do a new firmware compile as there are some interesting commits like https://git.lede-project.org/?p=source.git;a=commit;h=fa6fd45613d05da96b6733b15f91e51482378bb0
Unfortunately I ended up with a router that was only up and running for a few minutes and then it rebooted. After this happend a few times I decided to sysupgrade (downgrade) to the last build that was running perfectly. I guess the router decided to reboot in the middle of the upgrade procedure which resulted in a router that just went into a constant boot-loop. After a bit of soldering the serial console showed this.
[ 2.577402] # MTK NAND # : Use HW ECC
[ 2.581079] Device not found, ID: c8d1
[ 2.584805] Not Support this Device![ 2.588530] chip_mode=00000001
[ 2.591596] Support this Device in MTK table! c8d1[ 2.596544] [NAND]select ecc bit:4, sparesize :64 spare_per_sector=16
[ 2.603019] nand: device found, Manufacturer ID: 0xc8, Chip ID: 0xd1
[ 2.609339] nand: ESMT NAND 128MiB 3,3V 8-bit
[ 2.613691] nand: 128 MiB, SLC, erase size: 128 KiB, page size: 2048, OOB size: 64
[ 2.621254] Scanning device for bad blocks
[ 2.771199] 10 ofpart partitions found on MTD device MT7621-NAND
[ 2.777175] Creating 10 MTD partitions on "MT7621-NAND":
[ 2.782485] 0x000000000000-0x000000080000 : "Bootloader"
[ 2.789219] 0x000000080000-0x0000000c0000 : "Config"
[ 2.795413] 0x0000000c0000-0x000000100000 : "Bdata"
[ 2.801481] 0x000000100000-0x000000140000 : "Factory"
[ 2.807656] 0x000000140000-0x000000180000 : "crash"
[ 2.813968] 0x000000180000-0x0000001c0000 : "crash_syslog"
[ 2.820600] 0x0000001c0000-0x000000200000 : "reserved0"
[ 2.826888] 0x000000200000-0x000000600000 : "kernel_stock"
[ 2.833538] 0x000000600000-0x000000a00000 : "kernel"
[ 2.839547] 0x000000a00000-0x000007f80000 : "ubi"
[ 2.846300] [mtk_nand] probe successfully!
[ 2.851166] Signature matched and data read!
[ 2.855414] load_fact_bbt success 1023
[ 2.859683] libphy: Fixed MDIO Bus: probed
[ 2.933067] libphy: mdio: probed
[ 4.335817] mtk_soc_eth 1e100000.ethernet: loaded mt7530 driver
[ 4.342491] mtk_soc_eth 1e100000.ethernet eth0: mediatek frame engine at 0xbe100000, irq 20
[ 4.352719] NET: Registered protocol family 10
[ 4.358638] NET: Registered protocol family 17
[ 4.363185] 8021q: 802.1Q VLAN Support v1.8
[ 4.370043] UBI: auto-attach mtd9
[ 4.373428] ubi0: attaching mtd9
[ 5.486107] ubi0: scanning is finished
[ 5.505873] ubi0: attached mtd9 (name "ubi", size 117 MiB)
[ 5.511401] ubi0: PEB size: 131072 bytes (128 KiB), LEB size: 126976 bytes
[ 5.518241] ubi0: min./max. I/O unit sizes: 2048/2048, sub-page size 2048
[ 5.525012] ubi0: VID header offset: 2048 (aligned 2048), data offset: 4096
[ 5.531956] ubi0: good PEBs: 940, bad PEBs: 0, corrupted PEBs: 0
[ 5.537932] ubi0: user volume: 0, internal volumes: 1, max. volumes count: 128
[ 5.545136] ubi0: max/mean erase counter: 4/2, WL threshold: 4096, image sequence number: 719383500
[ 5.554156] ubi0: available PEBs: 916, total reserved PEBs: 24, PEBs reserved for bad PEB handling: 20
[ 5.563454] ubi0: background thread "ubi_bgt0d" started, PID 353
[ 5.563475] hctosys: unable to open rtc device (rtc0)
[ 5.574963] VFS: Cannot open root device "(null)" or unknown-block(0,0): error -6
[ 5.582439] Please append a correct "root=" boot option; here are the available partitions:
[ 5.590800] 1f00 512 mtdblock0 [ 5.594791] (driver?)
[ 5.597143] 1f01 256 mtdblock1 [ 5.601149] (driver?)
[ 5.603503] 1f02 256 mtdblock2 [ 5.607485] (driver?)
[ 5.609834] 1f03 256 mtdblock3 [ 5.613834] (driver?)
[ 5.616187] 1f04 256 mtdblock4 [ 5.620169] (driver?)
[ 5.622542] 1f05 256 mtdblock5 [ 5.626526] (driver?)
[ 5.628876] 1f06 256 mtdblock6 [ 5.632875] (driver?)
[ 5.635228] 1f07 4096 mtdblock7 [ 5.639210] (driver?)
[ 5.641583] 1f08 4096 mtdblock8 [ 5.645568] (driver?)
[ 5.647917] 1f09 120320 mtdblock9 [ 5.651917] (driver?)
[ 5.654269] Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0)
[ 5.664701] Rebooting in 1 seconds..
Using uboot console I was able to boot the stock kernel and then flash the stock firmware using the recovery mode of the stock kernel. After this I have installed my last working LEDE build and I'm up and running again.
Yikes! I just received mine tonight and I was going to attempt to load the latest snapshot. I guess I need to hold off?
Edit - and it sounds like the post above is a similar issue to what you are describing.....
Well I might just have been unlucky, but I would just hold off a few days until others have tested the latest snapshots with succes.
... unless you have the tools needed to gain UART access then just go ahead 
and report your findings.
I must say that everything is ok for me with current HEAD and wifi performance is getting close to 300 (there was a wifi driver update yesterday). I compile images myself, but it does not differ much from snapshots.
Maybe you should run a dirclean, remove .config and recompile from scratch?
2 Likes
Good advice, but after I followed it my router has boot loop
Wow! 300Mb of wifi download speed would be a dream for me!
Would be great if you can share the files to update to it from my current installed build.
If you are having bootloop issues than report them as bug.
It could be due to 4K sector change
Not really, those 4K changes are related to NOR, but this device has NAND that is unaffected
FYI I was able to solve the issue of losing access to internet (again to clarify, wifi connection was not dropped, simply I was unable to navigate) by simply unchecking the WMM setting in the Xiaomi wifi radio settings.
Unfortunately this doesn't help to increase the wifi download speed.
I have not yet updated because it is very stable
Compiled after a dirclean and updated without problems using sysupgrade.tar package.
So it sounds like latest snapshot is OK? Can I just follow the procedure in the OP versus compiling a new image? I just received the router and don't want to brick it. Finally, why are some of you compiling new images - to add fastpath? Thanks.
Well it works but I'm not sure about stability. My old build from the 6th of November (r5276-09f90b7) was running perfect for 10+ days. I did a new build after a dirclean (r5388-6fcf422) and after sysupgrading my router was rebooting every ~5 min (just as my previous story before I messed up everything).
I then did a factory reset and then it became stable (wierd...!). I did setup everything manually as it was before (of course there are probably minor changes) and I have experienced two reboots within the last 24 hours. Of course you might not experience the same as me as it could be a combination of packages installed and how everything is configured but at least my experience is that the new builds are not as stable as the "old" builds.
Yes
I compile because I need FPU emulation (make kernel_menuconfig - kernel type) for run MLdonkey on router and to directly add all the software I need, once learned it is very convenient and simple.
@hanner I was running r5117 for 26 days with many clients and without problems, I'll let you know if also with r5389 is stable.