FAQ? When N processes accessing transactional shared memory BDB and one process crashes?

Received	by 10.224.219.144 with SMTP id hu16mr4374532qab.1.1353129763511; Fri, 16 Nov 2012 21:22:43 -0800 (PST)
Received	by 10.182.192.97 with SMTP id hf1mr217756obc.36.1353129763358; Fri, 16 Nov 2012 21:22:43 -0800 (PST)
Path	csiph.com!usenet.pasdenom.info!news.franciliens.net!news.muarf.org!nntpfeed.proxad.net!proxad.net!feeder1-2.proxad.net!209.85.216.87.MISMATCH!u2no3938134qal.0!news-out.google.com!gf5ni327qab.0!nntp.google.com!u2no3960930qal.0!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail
Newsgroups	comp.databases.berkeley-db
Date	Fri, 16 Nov 2012 21:22:43 -0800 (PST)
Complaints-To	groups-abuse@google.com
Injection-Info	glegroupsg2000goo.googlegroups.com; posting-host=173.70.132.57; posting-account=KTOdKQoAAABD0Zlci66k82JaPtmbSuDj
NNTP-Posting-Host	173.70.132.57
User-Agent	G2/1.0
MIME-Version	1.0
Message-ID	<fbd1ac4e-26c2-4e44-b60e-85c6693e34a9@googlegroups.com> (permalink)
Subject	FAQ? When N processes accessing transactional shared memory BDB and one process crashes?
From	gshanemiller84@gmail.com
Injection-Date	Sat, 17 Nov 2012 05:22:43 +0000
Content-Type	text/plain; charset=ISO-8859-1
Xref	csiph.com comp.databases.berkeley-db:16

Show key headers only | View raw

I've heard the following:

- a single process with N threads accesses a transactional shared memory BDB and process crashes? No prob. run recover the restart application. BDB will be stable and accessible.
- N threads in M processes accessing a transactional shared memory BDB and one process crashes? the other M-1 processes are at risk, possibly hung. Running recover won't help or may be impossible because the M-1 processes running the other threads are hung

which if any is true?

Back to comp.databases.berkeley-db | Previous | Next | Find similar

Thread

FAQ? When N processes accessing transactional shared memory BDB and one process crashes? gshanemiller84@gmail.com - 2012-11-16 21:22 -0800

csiph-web