Re: [PATCH] cfq: fix starvation of asynchronous writes

2016-09-23 Thread Jens Axboe
On 09/22/2016 08:10 PM, Jens Axboe wrote: On 09/22/2016 06:59 PM, Glauber Costa wrote: While debugging timeouts happening in my application workload (ScyllaDB), I have observed calls to open() taking a long time, ranging everywhere from 2 seconds - the first ones that are enough to time out my

Re: [PATCH] cfq: fix starvation of asynchronous writes

2016-09-23 Thread Jens Axboe
On 09/22/2016 08:10 PM, Jens Axboe wrote: On 09/22/2016 06:59 PM, Glauber Costa wrote: While debugging timeouts happening in my application workload (ScyllaDB), I have observed calls to open() taking a long time, ranging everywhere from 2 seconds - the first ones that are enough to time out my

Re: [PATCH] cfq: fix starvation of asynchronous writes

2016-09-23 Thread Glauber Costa
On Fri, Sep 23, 2016 at 7:28 AM, Glauber Costa wrote: > On Sep 23, 2016 6:22 AM, "Paolo Valente" wrote: >> >> >> > Il giorno 23 set 2016, alle ore 02:59, Glauber Costa >> > ha scritto: >> > >> > While debugging timeouts

Re: [PATCH] cfq: fix starvation of asynchronous writes

2016-09-23 Thread Glauber Costa
On Fri, Sep 23, 2016 at 7:28 AM, Glauber Costa wrote: > On Sep 23, 2016 6:22 AM, "Paolo Valente" wrote: >> >> >> > Il giorno 23 set 2016, alle ore 02:59, Glauber Costa >> > ha scritto: >> > >> > While debugging timeouts happening in my application workload >> > (ScyllaDB), I have >> > observed

Re: [PATCH] cfq: fix starvation of asynchronous writes

2016-09-23 Thread Paolo Valente
> Il giorno 23 set 2016, alle ore 02:59, Glauber Costa > ha scritto: > > While debugging timeouts happening in my application workload (ScyllaDB), I > have > observed calls to open() taking a long time, ranging everywhere from 2 > seconds - > the first ones that are

Re: [PATCH] cfq: fix starvation of asynchronous writes

2016-09-23 Thread Paolo Valente
> Il giorno 23 set 2016, alle ore 02:59, Glauber Costa > ha scritto: > > While debugging timeouts happening in my application workload (ScyllaDB), I > have > observed calls to open() taking a long time, ranging everywhere from 2 > seconds - > the first ones that are enough to time out my

Re: [PATCH] cfq: fix starvation of asynchronous writes

2016-09-22 Thread Jens Axboe
On 09/22/2016 06:59 PM, Glauber Costa wrote: While debugging timeouts happening in my application workload (ScyllaDB), I have observed calls to open() taking a long time, ranging everywhere from 2 seconds - the first ones that are enough to time out my application - to more than 30 seconds. The

Re: [PATCH] cfq: fix starvation of asynchronous writes

2016-09-22 Thread Jens Axboe
On 09/22/2016 06:59 PM, Glauber Costa wrote: While debugging timeouts happening in my application workload (ScyllaDB), I have observed calls to open() taking a long time, ranging everywhere from 2 seconds - the first ones that are enough to time out my application - to more than 30 seconds. The

[PATCH] cfq: fix starvation of asynchronous writes

2016-09-22 Thread Glauber Costa
While debugging timeouts happening in my application workload (ScyllaDB), I have observed calls to open() taking a long time, ranging everywhere from 2 seconds - the first ones that are enough to time out my application - to more than 30 seconds. The problem seems to happen because XFS may block

[PATCH] cfq: fix starvation of asynchronous writes

2016-09-22 Thread Glauber Costa
While debugging timeouts happening in my application workload (ScyllaDB), I have observed calls to open() taking a long time, ranging everywhere from 2 seconds - the first ones that are enough to time out my application - to more than 30 seconds. The problem seems to happen because XFS may block