> I found that subsequent reads on a small timescale often succeed. Also after > a pause of some couple of seconds reads probably succeed. THe timescale for > faillure lies inbetween. Please try a time distribution more randomly. > How big is your sample I used 1GB. When testing by hand 1 out of 6 reads fail. > Same behavior on stgt 0.8 and 0.9.0 . I use 1GB as well. It would be helpful if you provide me with a script that does these random timings between reads. Also, I noted that after one read, no I/O is going any more on the target side, as this 1GB probably gets cached. My backing store is an sdb block device and I wasn't sure what's your and if you have caching at all, maybe this influences something. Or. -- To unsubscribe from this list: send the line "unsubscribe stgt" in the body of a message to majordomo at vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html |