[mirror-admin] Still having problems syncing

[Chris Adams]

I am still seeing an incredibly slow file list fetch while trying to
rsync fedora-enchilada.  I have an rsync that has been running for over
2 hours and it is still waiting on the file list.  I attached with
strace, and I see bits trickling in every 30 seconds:

20:07:52 read(3, "\374\17\0\7", 4)      = 4
20:07:52 time(NULL)                     = 1247706472
20:07:52 select(4, [3], [], NULL, {60, 0}) = 1 (in [3], left {60, 0})
20:07:52 read(3, "\22D\4H@\330\35\1>\0062\36daemon-1.5.4.5-3.fc9"..., 4092) = 4040
20:07:52 time(NULL)                     = 1247706472
20:07:52 select(4, [3], [], NULL, {60, 0}) = 1 (in [3], left {59, 936000})
20:07:52 read(3, "\v\0\26F\4HB\354#\0>\0065\34devel-0.6.24-1.fc9"..., 52) = 52
20:07:52 time(NULL)                     = 1247706472
20:07:52 select(4, [3], [], NULL, {60, 0}) = 1 (in [3], left {28, 875000})
20:08:24 read(3, "\374\17\0\7", 4)      = 4
20:08:24 time(NULL)                     = 1247706504
20:08:24 select(4, [3], [], NULL, {60, 0}) = 1 (in [3], left {60, 0})
20:08:24 read(3, "#\0>\0065\33smsd-0.6.24-1.fc9.ppc64.rp"..., 4092) = 2692
20:08:24 time(NULL)                     = 1247706504
20:08:24 select(4, [3], [], NULL, {60, 0}) = 1 (in [3], left {59, 999000})
20:08:24 read(3, "4.rpm\240-=\0bP\17H\201q\205\0>\6:\36devel-2.22."..., 1400) = 1348
20:08:24 time(NULL)                     = 1247706504
20:08:24 select(4, [3], [], NULL, {60, 0}) = 1 (in [3], left {59, 939000})
20:08:24 read(3, "@\34pell-2.19.1-15.fc9.ppc64.rpm\331E"..., 52) = 52
20:08:24 time(NULL)                     = 1247706504
20:08:24 select(4, [3], [], NULL, {60, 0}) = 1 (in [3], left {45, 12000})
20:08:39 read(3, "\374\17\0\7", 4)      = 4
20:08:39 time(NULL)                     = 1247706519
20:08:39 select(4, [3], [], NULL, {60, 0}) = 1 (in [3], left {60, 0})
20:08:39 read(3, "-2.19.1-15.fc9.ppc64.rpm\225\r\1\0\246G\4H"..., 4092) = 2692
20:08:39 time(NULL)                     = 1247706519
20:08:39 select(4, [3], [], NULL, {60, 0}) = 1 (in [3], left {59, 999000})
20:08:39 read(3, "H\4H~:[\1>\0067$plugins-extras-1.8.2-"..., 1400) = 1348
20:08:39 time(NULL)                     = 1247706519
20:08:39 select(4, [3], [], NULL, {60, 0}) = 1 (in [3], left {59, 941000})
20:08:39 read(3, "\0PI\4H2\315;\1>\0060\31erf-3.0.3-4.fc9.ppc"..., 52) = 52
20:08:39 time(NULL)                     = 1247706519
20:08:39 select(4, [3], [], NULL, {60, 0}) = 1 (in [3], left {34, 832000})
20:09:04 read(3, "\374\17\0\7", 4)      = 4
20:09:04 time(NULL)                     = 1247706544
20:09:04 select(4, [3], [], NULL, {60, 0}) = 1 (in [3], left {60, 0})
20:09:04 read(3, "2\34iodic-2.0.10-3.fc9.ppc64.rpm\333\23"..., 4092) = 4040
20:09:04 time(NULL)                     = 1247706544
20:09:04 select(4, [3], [], NULL, {60, 0}) = 1 (in [3], left {59, 939000})
20:09:04 read(3, "9.ppc64.rpm\2526*\0\245j\33H\10iX\1>\0060\32k+-1."..., 52) = 52
20:09:04 time(NULL)                     = 1247706544

I understand that Red Hat has been looking at this; has there been any
progress on fixing this?  Cutting down the simultaneous connections
hasn't seemed to have helped.

-- 
Chris Adams <cmadams at hiwaay.net>
Systems and Network Administrator - HiWAAY Internet Services
I don't speak for anybody but myself - that's enough trouble.

--