Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 11 Mar 2014 08:29:33 -0500
From:      Karl Denninger <tickerguydenninger@gmail.com>
To:        freebsd-stable@freebsd.org
Subject:   Two odd problems with STABLE-10 r262921
Message-ID:  <CAHCMRk_=s%2B2LYr-pLkt7LJK3LcWSiomtLb_HhfUrj4VMUHjQVQ@mail.gmail.com>

next in thread | raw e-mail | index | archive | help
Two things I've run into with this coming from 9.2-STABLE....

1. I am getting errors coming from mail transmissions to certain MX relays
-- and only those relays.  One of them is (ironically) mx1.freebsd.org,
which precludes emailing the list from my primary email address!  The error
logs in the maillog file show:

Mar 11 08:17:46 NewFS sm-mta[3605]: STARTTLS=client, relay=mx1.freebsd.org.,
version=TLSv1/SSLv3, verify=FAIL, cipher=ECDHE-RSA-AES256-GCM-SHA384,
bits=256/256
Mar 11 08:17:46 NewFS sm-mta[3605]: STARTTLS: write error=syscall error
(-1), errno=13, get_error=error:00000000:lib(0):func(0):reason(0),
retry=99, ssl_err=5
Mar 11 08:17:46 NewFS sm-mta[3605]: s2AKht3B064414: SYSERR(root): putbody:
write error: Permission denied
Mar 11 08:17:46 NewFS sm-mta[3605]: s2AKht3B064414: SYSERR(root): timeout
writing message to mx1.freebsd.org.: Permission denied
Mar 11 08:17:46 NewFS sm-mta[3605]: s2AKht3B064414: to=<
freebsd-fs@freebsd.org>, ctladdr=<karl@denninger.net> (1001/1001),
delay=16:33:50, xdelay=00:00:05, mailer=esmtp, pri=4186247, relay=
mx1.freebsd.org. [8.8.178.115], dsn=4.0.0, stat=Deferred

Permission denied -- on a socket?  As root?  What am I missing here?

(Shutting off TLS does not resolve this.)  However, this is not universal;
it only impacts *some* emails....


Mar 11 08:20:37 NewFS sm-mta[5433]: s2BDKbF4005433: from=<
ticker@fs.denninger.net>, size=962, class=0, nrcpts=1, msgid=<
201403111320.s2BDKTF3005412@fs.denninger.net>, proto=ESMTP, daemon=IPv4,
relay=localhost [127.0.0.1]
Mar 11 08:20:37 NewFS sendmail[5412]: s2BDKTF3005412: to=xxxxxxxx@yahoo.com,
ctladdr=ticker (20098/20098), delay=00:00:08, xdelay=00:00:05,
mailer=relay, pri=3
0494, relay=[127.0.0.1] [127.0.0.1], dsn=2.0.0, stat=Sent (Message accepted)
Mar 11 08:20:37 NewFS sm-mta[5461]: STARTTLS=client, relay=
mta5.am0.yahoodns.net., version=TLSv1/SSLv3, verify=FAIL,
cipher=DHE-RSA-CAMELLIA256-SHA, bits=256/256
Mar 11 08:20:39 NewFS sm-mta[5461]: s2BDKbF4005433: to=<xxxxxxx@yahoo.com>,
ctladdr=<ticker@fs.denninger.net> (20098/20098), delay=00:00:02,
xdelay=00:00:02,
mailer=esmtp, pri=30962, relay=mta5.am0.yahoodns.net. [66.196.118.35],
dsn=2.0.0, stat=Sent (ok dirdel)

That one went through successfully....

This is new; I didn't have any trouble on 9.2-STABLE at all.  Ideas?

The other issue is that I am running into bizarre wired memory behavior.  I
have ZFS on the box and it was wiring down ~20GB, driving the system into
paging.  I dropped arc_max to 16GB as a hard limit, it's currently showing
15GB, but...

20531152 wire

zfs-stats -A shows:


------------------------------------------------------------------------
ZFS Subsystem Report                            Tue Mar 11 08:27:55 2014
------------------------------------------------------------------------

ARC Summary: (HEALTHY)
        Memory Throttle Count:                  0

ARC Misc:
        Deleted:                                4.22m
        Recycle Misses:                         57.10k
        Mutex Misses:                           320
        Evict Skips:                            2.35m

ARC Size:                               95.96%  15.35   GiB
        Target Size: (Adaptive)         100.00% 16.00   GiB
        Min Size (Hard Limit):          12.50%  2.00    GiB
        Max Size (High Water):          8:1     16.00   GiB

ARC Size Breakdown:
        Recently Used Cache Size:       75.49%  12.08   GiB
        Frequently Used Cache Size:     24.51%  3.92    GiB

ARC Hash Breakdown:
        Elements Max:                           1.59m
        Elements Current:               55.53%  883.80k
        Collisions:                             5.84m
        Chain Max:                              16
        Chains:                                 261.48k

------------------------------------------------------------------------

Ok, where's the rest coming from?  There's nothing in the shared memory
segment list that appears to be related.

Is there a memory leak in here somewhere?



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAHCMRk_=s%2B2LYr-pLkt7LJK3LcWSiomtLb_HhfUrj4VMUHjQVQ>