[sf-lug] should be "all better" again within about 8 hours: Re: BALUG.ORG [www.]balug.org temporarily down (use new.balug.org as partial work-around); meeting still happening this Tuesday

[Michael Paoli]

Presently guestimating this should be "all better"* again, in a bit  
under 8 hours.

Billing account there had been suspended (billing glitch), and has
been unsuspended (thanks Michael Hubbard).
 From DreamHost.com support:
Hoffman: at 21:59:04
Account was unsuspended about 20 minutes ago

Now we're mostly waiting on DNS.
DreamHost.com uses 3 DNS servers, but they're indirect, and in some type of
load balancing ring or the like - e.g. one can query the same IP repeatedly,
and get various (including older) SOA serial numbers for the same zone - but
they do eventually all catch up.  So, we essentially have 2 TTLs to  
wait through
for DNS to be "all better"* again.

$ (for ns in $(dig -t NS balug.org. +short); do echo "$ns" $(dig  
@"$ns" -t SOA balug.org. +short); done)
ns1.dreamhost.com. ns1.dreamhost.com. hostmaster.dreamhost.com.  
2016021400 17454 1800 1814400 14400
ns3.dreamhost.com. ns1.dreamhost.com. hostmaster.dreamhost.com.  
2016021401 18977 1800 1814400 14400
ns2.dreamhost.com. ns1.dreamhost.com. hostmaster.dreamhost.com.  
2016021401 18977 1800 1814400 14400
$ (for ns in $(dig -t NS balug.org. +short); do echo "$ns":; dig  
@"$ns" -t SOA balug.org. | grep '^[    ]*[^    ;]'; done)
ns3.dreamhost.com.:
balug.org.              14400   IN      SOA     ns1.dreamhost.com.  
hostmaster.dreamhost.com. 2016021401 18977 1800 1814400 14400
ns2.dreamhost.com.:
balug.org.              14400   IN      SOA     ns1.dreamhost.com.  
hostmaster.dreamhost.com. 2016021401 18977 1800 1814400 14400
ns1.dreamhost.com.:
balug.org.              14400   IN      SOA     ns1.dreamhost.com.  
hostmaster.dreamhost.com. 2016021401 18977 1800 1814400 14400
$

We're looking at up to TTL for zone (for remaining NS to pick up updates) +
negative caching TTL (where DNS elsewhere has cached that A records  
don't exist),
both of which happen to be 4 hours - so, up to 8 hours total, relative to when
the DNS corrections were first made at their point of origin (whatever
"feeds" DreamHost.com's DNS nameservers).

Serial 2016021401 has the A records for [www.]balug.org., whereas  
2016021400 does not.

*well, within the limits of DreamHost.com typical behavior.

> From: "Michael Paoli" <Michael.Paoli at cal.berkeley.edu>
> Subject: BALUG.ORG [www.]balug.org temporarily down (use  
> new.balug.org as partial work-around); meeting still happening this  
> Tuesday
> Date: Sun, 14 Feb 2016 21:22:48 -0800

> There's been a hosting SNAFU with [www.]balug.org (not sure exactly why
> yet - hoping to hear back on that soon), Presently down:
> [www.]balug.org and BALUG lists & email
> (partial) work-around:
> http://www.new.balug.org/ (that host/site has essentially everything except
> lists/email) and doesn't have www.balug.org and balug.org itself - but does
> have everything else).
>
> Meetings go do on per usual - so yes, will have our regular meeting
> 3rd Tuesday of the month - this coming Tuesday.
>
> I'm guestimating the [www.]balug.org stuff will be sorted out 'n  
> fixed (or more
> fully worked around) in 48 to 96 hours - possibly more, or less.
> In the meantime, http://www.new.balug.org/ can be used as at least
> partial work-around.
>
> Sorry for the inconvenience.