On Fri, 19 Jan 96, Peter Seebach <seebs@solon.com> wrote:

>Since about the 7th, I have had the *majority* of sup transfers look like:
>SUP 8.26 (4.3 BSD) for file /etc/supfile at Jan 19 22:49:42
>SUP Upgrade of current-allsrc at Fri Jan 19 22:49:49 1996
>SUP Fileserver 8.13 (4.3 BSD) 23099 on netbsd.warped.com at 22:49:49
>SUP Requesting changes since Jan 19 07:33:25 1996
>SUP Using compressed file transfer
>SUP: Read error on network: Operation timed out
>SUP: Premature EOF on network input
>SUP: Network write timed out
>SUP: Error receiving file from file server
>SUP: Upgrade of current-allsrc aborted at Jan 19 22:59:41 1996
>SUP: Network write timed out
>SUP: Aborted
>
>... This seems to happen entirely too often.  Am I missing something
>obvious?

FYI, if anyone cares...

I tried to sup for the first time since early October, last night.  I
first extracted a stock set of 1.1 sources on top of my (October
current) sources, so I wouldn't have to re-fetch every file.

I made it all the way through allsource, and was getting
current-security when I finally got this:

SUP Receiving file src/domestic/lib/libcrypt/crypt.3
SUP Receiving file src/domestic/lib/libcrypt/crypt.c
SUP: Read error on network: Connection reset by peer
SUP: Premature EOF on network input
SUP: Network write timed out
SUP: Error receiving file from file server
SUP: Upgrade of current-security aborted at Jan 20 00:07:54 1996
SUP: Network write timed out
SUP: Aborted

My sup had started at 20:50:04, so this was 3:17:50, more than 3
hours, later (it still fetched several files, and had to "update" them
all).  I then got:

SUP Receiving file othersrc/cvs/lib/hostname.c
SUP Receiving file othersrc/cvs/lib/md5.c
SUP: Read error on network: Connection reset by peer
SUP: Premature EOF on network input
SUP: Network write timed out
SUP: Error receiving file from file server
SUP: Upgrade of current-othersrc aborted at Jan 20 00:25:12 1996
SUP: Network write timed out
SUP: Aborted

This being 00:17:18 later.  Those were the only two aborts I saw.

Here is a traceroute from my host to sup.netbsd.org:

[michaelv@MindBender]~> traceroute sup.netbsd.org
traceroute to netbsd.warped.com (205.149.163.23), 30 hops max, 40 byte packets
 1  clown.seanet.com (199.181.164.108)  168.191 ms  191.53 ms  168.922 ms
 2  roto.seanet.com (199.181.164.100)  157.394 ms  154.957 ms  157.586 ms
 3  sl-stk-11-S1/5-T1.sprintlink.net (144.228.141.37)  177.753 ms  174.366 ms  157.834 ms
 4  sl-stk-6-F0/0.sprintlink.net (144.228.40.6)  247.977 ms  174.389 ms  167.776 ms
 5  sl-stk-nap-H2/0-T3.sprintlink.net (144.228.10.50)  177.424 ms  244.329 ms  187.71 ms
 6  pb-nap.agis.net (198.32.128.19)  177.528 ms  195.654 ms  167.803 ms
 7  mae-west.best.net (198.32.136.36)  197.649 ms  184.724 ms  177.522 ms
 8  netbsd.warped.com (205.149.163.23)  187.918 ms  185.092 ms  177.58 ms

Sometimes, when Seanet is connected directly to its agis link, I am
only six hops away (much of last night).

Here is my supfile:

current release=include host=sup.NetBSD.org hostbase=/ftp/pub base=/usr prefix=/usr backup delete use-rel-suffix old
current release=src host=sup.NetBSD.org hostbase=/ftp/pub base=/usr prefix=/usr backup delete use-rel-suffix
current release=security host=sup.NetBSD.org hostbase=/ftp/pub base=/usr prefix=/usr delete use-rel-suffix old
current release=ksrc-common host=sup.NetBSD.org hostbase=/ftp/pub base=/usr prefix=/usr backup delete use-rel-suffix
current release=ksrc-i386 host=sup.NetBSD.org hostbase=/ftp/pub base=/usr prefix=/usr backup delete use-rel-suffix
#current release=gnu host=sup.NetBSD.org hostbase=/ftp/pub base=/usr prefix=/usr backup delete use-rel-suffix
current release=othersrc host=sup.NetBSD.org hostbase=/ftp/pub base=/usr prefix=/usr backup use-rel-suffix
current release=doc host=sup.NetBSD.org hostbase=/ftp/pub base=/usr prefix=/usr backup use-rel-suffix old
current release=games host=sup.NetBSD.org hostbase=/ftp/pub base=/usr prefix=/usr use-rel-suffix old

As you can see, my order of supping goes: include, src, security,
ksrc-common, ksrc-i386, othersrc, doc, games.

Finally, if it matters, I also would like to plead that the src
distribution be split into smaller separate distributions.  Although I
didn't have any trouble with it this time, I have, occasionally, in
the past.

-----------------------------------------------------------------------------
  Michael L. VanLoon                                 michaelv@HeadCandy.com
       --<  Free your mind and your machine -- NetBSD free un*x  >--
     NetBSD working ports: 386+PC, Mac 68k, Amiga, HP300, Sun3, Sun4,
                           DEC PMAX (MIPS), DEC Alpha, PC532
     NetBSD ports in progress: VAX, Atari 68k, others...
-----------------------------------------------------------------------------