What's new

USB3 bug or heavy load hanging AC87U - 380.68_2

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

fry

Occasional Visitor
Hey guys,

Not sure if this is a bug, or my hard drive is about to die. I've got over 100GB of files I'm trying to move from my Macbook to my USB3 drive shared on my AC87U over ethernet. The same Macbook is downloading files at 4MB/s. I've got 14 other wireless clients, but nothing else noteworthy using bandwidth.

Last night I started moving the files and noticed that after 5-10mins, the router appeared to crash/hang. I couldn't manage it wireless or over ethernet so turned it off/on.

The file copy started again (it's an automated job), and it happened again, 5-10 mins later. Turned it off/on and dug into the logs, finding this;

Sep 29 09:12:25 kernel: xhci_hcd 0000:00:0c.0: WARN: Stalled endpoint
Sep 29 09:12:25 kernel: sd 0:0:0:0: [sda] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
Sep 29 09:12:25 kernel: sd 0:0:0:0: [sda] Sense Key : Illegal Request [current]
Sep 29 09:12:25 kernel: sd 0:0:0:0: [sda] Add. Sense: Invalid command operation code
Sep 29 09:12:25 kernel: sd 0:0:0:0: [sda] CDB: Write same(16): 93 08 00 00 00 00 6c 7c 45 78 00 00 01 d8 00 00
Sep 29 09:12:25 kernel: end_request: I/O error, dev sda, sector 1820083576
Sep 29 09:12:25 kernel: end_request: I/O error, dev sda, sector 1820083576

Doing a quick google, it would suggest there is either a problem with the disk https://vip.asus.com/forum/view.asp...R&id=20141224042950486&page=1&SLanguage=en-us, or a USB3/2 bug -> https://ubuntuforums.org/archive/index.php/t-1642641.html

Not really sure where to start...
 
And again

Sep 29 11:53:03 kernel: xhci_hcd 0000:00:0c.0: WARN: Stalled endpoint
Sep 29 11:53:03 kernel: sd 0:0:0:0: [sda] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
Sep 29 11:53:03 kernel: sd 0:0:0:0: [sda] Sense Key : Illegal Request [current]
Sep 29 11:53:03 kernel: sd 0:0:0:0: [sda] Add. Sense: Invalid command operation code
Sep 29 11:53:03 kernel: sd 0:0:0:0: [sda] CDB: Write same(16): 93 08 00 00 00 00 76 b4 15 30 00 01 f6 e8 00 00
Sep 29 11:53:03 kernel: end_request: I/O error, dev sda, sector 1991513392
Sep 29 11:53:03 kernel: end_request: I/O error, dev sda, sector 1991513392
 
Pulled it out and plugged into my Macbook, ran First Aid and found;

Checking prerequisites
Checking the partition list
Problems were found with the partition map which might prevent booting
Operation successful.

Nothing really interesting yet.
 
Ran Paragon NTFS for Mac and found no issues/bad sectors.

Verifying file system.
Checking Volume /dev/rdisk2s1...
Type of the filesystem is NTFS.
Volume label is: <snip>.
$UpCase file is formatted for use in Windows 7 and later versions.
3725.99 GB total disk space.
1042.65 GB in 2699 files.
964 KB in 814 directories.
210464 KB in use by the system.
64 MB occupied by the log/journal file.
4 KB in each allocation unit.
976745727 total allocation units on volume (3725.99 GB).
703367353 allocation units available on volume (2683.13 GB).
No repairs were necessary for volume /dev/rdisk2s1.
File system check exit code is 0.
 
Connected it back and still have errors when writing to the drive.

Sep 29 15:17:38 kernel: end_request: I/O error, dev sda, sector 347650336
Sep 29 15:17:38 kernel: end_request: I/O error, dev sda, sector 347650336
Sep 29 15:17:38 kernel: end_request: I/O error, dev sda, sector 671694688
Sep 29 15:17:38 kernel: end_request: I/O error, dev sda, sector 671694688
Sep 29 15:19:21 kernel: end_request: I/O error, dev sda, sector 671694784
Sep 29 15:19:21 kernel: end_request: I/O error, dev sda, sector 671694784
Sep 29 15:19:21 kernel: end_request: I/O error, dev sda, sector 671696256
Sep 29 15:19:21 kernel: end_request: I/O error, dev sda, sector 671696256
Sep 29 15:19:21 kernel: end_request: I/O error, dev sda, sector 671697040
Sep 29 15:19:21 kernel: end_request: I/O error, dev sda, sector 671697040
Sep 29 15:19:21 kernel: end_request: I/O error, dev sda, sector 671697656
Sep 29 15:19:21 kernel: end_request: I/O error, dev sda, sector 671697656
Sep 29 15:19:50 kernel: end_request: I/O error, dev sda, sector 671694648
Sep 29 15:19:50 kernel: end_request: I/O error, dev sda, sector 671694648
Sep 29 15:19:50 kernel: end_request: I/O error, dev sda, sector 678535560
Sep 29 15:19:50 kernel: end_request: I/O error, dev sda, sector 678535560
Sep 29 15:19:56 kernel: end_request: I/O error, dev sda, sector 1717673752
Sep 29 15:19:56 kernel: end_request: I/O error, dev sda, sector 1717673752
Sep 29 15:19:56 kernel: end_request: I/O error, dev sda, sector 684040600
Sep 29 15:19:56 kernel: end_request: I/O error, dev sda, sector 684040600
Sep 29 15:20:08 kernel: end_request: I/O error, dev sda, sector 687904336
Sep 29 15:20:08 kernel: end_request: I/O error, dev sda, sector 687904336
Sep 29 15:22:39 kernel: end_request: I/O error, dev sda, sector 1393893736
Sep 29 15:22:39 kernel: end_request: I/O error, dev sda, sector 1393893736
Sep 29 15:22:39 kernel: end_request: I/O error, dev sda, sector 1440840888
Sep 29 15:22:39 kernel: end_request: I/O error, dev sda, sector 1440840888
Sep 29 15:22:39 kernel: end_request: I/O error, dev sda, sector 680037184
Sep 29 15:22:39 kernel: end_request: I/O error, dev sda, sector 680037184
Sep 29 15:22:39 kernel: end_request: I/O error, dev sda, sector 684040544
Sep 29 15:22:39 kernel: end_request: I/O error, dev sda, sector 684040544
Sep 29 15:22:39 kernel: end_request: I/O error, dev sda, sector 1444678648
Sep 29 15:22:39 kernel: end_request: I/O error, dev sda, sector 1444678648
Sep 29 15:22:39 kernel: end_request: I/O error, dev sda, sector 678535520
Sep 29 15:22:39 kernel: end_request: I/O error, dev sda, sector 678535520
Sep 29 15:22:39 kernel: end_request: I/O error, dev sda, sector 1444678760
Sep 29 15:22:39 kernel: end_request: I/O error, dev sda, sector 1444678760
Sep 29 15:24:38 kernel: end_request: I/O error, dev sda, sector 1455100240
Sep 29 15:24:38 kernel: end_request: I/O error, dev sda, sector 1455100240
Sep 29 15:24:38 kernel: end_request: I/O error, dev sda, sector 1456099464
Sep 29 15:24:38 kernel: end_request: I/O error, dev sda, sector 1456099464
 
Those errors often indicate a problem with the USB cable or the HDD not getting sufficient power.

Also make sure you disabled the USB 3.0 interference option under Wireless -> Professional.
 
Thanks Merlin, the Interference option is already disabled (default). I can't say I haven't seen those logs before, but haven't had this issue so haven't been stalking my log file.
 
87U here, on 68_2 with a 4TB seagate desktop backup+ (i.e., self powered) mounted as NTFS on the USB3 port.

I started an 80Gb transfer from a laptop over the 5Gz connection to the seagate drive, and it is banging away just fine. It is throttled by the wifi speed, and the router is ranging between 20-100% on the two cores--alternating. No log messages.

So, this suggests it must be the drive or the cable as Merlin suggested.

For giggles, I also started a 100GB transfer from a FREENAS server to the seagate drive over wired connections, which moves the router to 80-95% on both cores. Will see.
 
So I can replace the cable (pretty easy but strange I haven't had this issue before, same cable for 7 months) but if it's the drive and I can't find any bad sectors, what is one to do?
 
You didn't specify how your hard disk is obtaining its power. If it's USB-powered, then the disk simply draws too much current for the router's USB port. You'd have to either connect an external power supply to the hard disk if it supports it, or otherwise replace it with a hard disk that comes with its own power supply.

Those issues are quite common, especially with Seagate disks. I've seen some of these Seagate USB-powered disk fail even on a desktop computer, until we hooked them to an external power supply, which solved the issue.
 
You didn't specify how your hard disk is obtaining its power. If it's USB-powered, then the disk simply draws too much current for the router's USB port. You'd have to either connect an external power supply to the hard disk if it supports it, or otherwise replace it with a hard disk that comes with its own power supply.

Those issues are quite common, especially with Seagate disks. I've seen some of these Seagate USB-powered disk fail even on a desktop computer, until we hooked them to an external power supply, which solved the issue.
It's a Western Digital 4TB 2.5" USB3 drive powered by the USB cable. If it's drawing more than the USB3 interface allows, surely that's a manufacturing/design fault.

* Does the USB port on the router not provide full current?
* If it was power related, wouldn't I be getting more verbose error messages about power being insufficient?
 
Last edited:
* If it was power related, wouldn't I be getting more verbose error messages about power being insufficient?

No, the kernel has no way of knowing what is happening if electrical signals within the HDD or its enclosure aren't up to specs.

A 4 TB HDD is probably pushing the specs to the limit, especially if it's actually a pair of 2 TB HDDs within the enclosure.

It's commonly known in the field that USB-powered HDDs aren't always as stable.
 
No, the kernel has no way of knowing what is happening if electrical signals within the HDD or its enclosure aren't up to specs.

A 4 TB HDD is probably pushing the specs to the limit, especially if it's actually a pair of 2 TB HDDs within the enclosure.

It's commonly known in the field that USB-powered HDDs aren't always as stable.
Interesting, not something I really thought of but makes sense. Thanks for schooling me :)
 
If I recall correctly, USB3 hubs aren't supported. He could try a USB3 Y cable, but those are pretty pricy.

He might plug in a smartphone that is capable of drawing more than 900ma and see if his USB3 port is supplying 900ma with an app like Ampere.
 

Similar threads

Latest threads

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top