Re: [Nbd] NBD Super Size io operations
- To: Wouter Verhelst <w@...112...>
- Cc: nbd-general@lists.sourceforge.net
- Subject: Re: [Nbd] NBD Super Size io operations
- From: Paul Clements <paul.clements@...856...>
- Date: Tue, 9 Apr 2013 12:36:35 -0400
- Message-id: <CAECXXi6BUf9B-Cv9odRFY+LvKE6UcF1e+SSMOqjA+u-JuZpyag@...18...>
- In-reply-to: <51643A3E.1050409@...112...>
- References: <BAY403-EAS415E8D3A1CC43EB4134B824DADE0@...59...> <CAECXXi5ouWsq09pxTGzDeYP+Q6eLF4+SbBRDKShVcUXm13Oi4g@...18...> <20130401202523.GR14908@...1273...> <20130402140706.GI30504@...855...> <20130403193632.GY30504@...855...> <20130407215919.GB67997@...1273...> <CAECXXi5fiTnfgcHhmW_A84nwXUhvvCX_CZf1QLzydSG09hZ6Pw@...18...> <CAECXXi59Zea8BVLER3S9Kf=uJ3pikuUxzO1QG_wQUVQqs6axAg@...18...> <51643A3E.1050409@...112...>
On Tue, Apr 9, 2013 at 11:56 AM, Wouter Verhelst <w@...112...> wrote:
> On 08-04-13 23:19, Paul Clements wrote:
>> If you do something like the following in nbd-client just before the
>> call to NBD_DO_IT, does this fix the problem:
>>
>>
>> + sigfillset(&block);
>> + sigdelset(&block, SIGKILL);
>> + sigprocmask(SIG_SETMASK, &block, &old);
>> if (ioctl(nbd, NBD_DO_IT) < 0)
>
> Hmm. I'm not sure if blocking everything except for SIGKILL is the right
> thing to do. Shouldn't we at least listen for SIGINT and SIGTERM, too?
> And what about SIGPIPE?
Yeah, you're right. For my version of nbd-client, just SIGKILL is
fine, but generally whatever signals you don't mind aborting the nbd
connection on, you can allow. Note that sometimes nbd-client will
block all but SIGKILL anyway (if you're in a network send/recv,
NBD_DO_IT masks all but SIGKILL).
--
Paul
Reply to: