Hi On 11.06.2015 07:46, Axel Christiansen wrote: > Hello List. > > > I am having trouble solving a problem with a bunch of USB3-SATA-bridge connected to 2 USB3-hubs. I was not able finding any useful help for weeks now. > > Perhaps by bringing this issue to linux-usb can help me troubleshot the problem or sort out a bug or help improve Linux USB stability. > > The issue is an ongoing reset of different JMicron (ID 152d:0567) JMS567 USB3-SATA-bridges partitioned equally to 2 USB3-hubs (ID 2109:0812). > > I have 8 of theses bridges with 8 3TB WD (WDC WD30EZRX) hard drives connected to them. That 8 drives are configured running as a raid6. > > Reading/writing using 1 hard disk at a time works fine, even with higher speeds using dd or copying large files. Using several drives simultaneously is causing ongoing resets on a number of these bridges randomly (but 1 a time). It is like reading for seconds, then the read blocks for may by 5 seconds, a reset on one bridge happens and the reading continues. > > The host is a Debian 8 with Linux 3.16.0-4-amd64. > > What i noticed today was, by lowering max_sectors to 16 or 8 improves stability much, but it is still fare from stable. > > > What could be done improving this? > Are there any quirks worth trying? > > > Any idea would be much appreciated. Axel > At least xhci has several improvements since 3.16, Can you try to reproduce this with a more recent kernel? 4.1-rc7 preferably, or 4.1 final once it's out. -Mathias -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
