You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Jim Jagielski <ji...@jaguNET.com> on 2016/11/05 22:24:05 UTC
mod_socache_redis
Trunk now includes mod_socache_redis, which requires APR 1.6
and above. Please test it out and provide feedback :)
I also added in the required mod_status hook info for mod_socache_memcache
as well, so now it was also provide mod_status info. I plan to
also submit that for backporting to 2.4 as well ;)
Re: mod_socache_redis
Posted by Hank Ibell <hw...@gmail.com>.
Hello again Jim,
Thanks for the quick response. All is well now with the latest updates.
Thanks!
Regards,
Hank Ibell
On Sun, Nov 6, 2016 at 3:21 PM, Jim Jagielski <ji...@jagunet.com> wrote:
> Somehow things got out of sync. Should be fixed
> now.
>
> > On Nov 6, 2016, at 2:59 PM, Jacob Champion <ch...@gmail.com> wrote:
> >
> > On 11/06/2016 05:32 AM, Jim Jagielski wrote:
> >> This is weird... Are you using the stuff at the ASF??
> >> The format warning are way wrong as well:
> >>
> >> /** Keyspace hits */
> >> apr_uint64_t keyspace_hits;
> >> /** Keyspace misses */
> >> apr_uint64_t keyspace_misses;
> >>
> >> keyspace_* are obviously not apr_uint32_ts
> >
> > They're uint32s in my local copy too, and the version on the server:
> >
> > http://svn.apache.org/viewvc/apr/apr-util/branches/1.6.x/
> include/apr_redis.h?view=markup#l420
> >
> > Same for the version in APR 2.0. Did we miss some alternate instructions
> for the new code?
> >
> > --Jacob
>
>
Re: mod_socache_redis
Posted by Jim Jagielski <ji...@jaguNET.com>.
Somehow things got out of sync. Should be fixed
now.
> On Nov 6, 2016, at 2:59 PM, Jacob Champion <ch...@gmail.com> wrote:
>
> On 11/06/2016 05:32 AM, Jim Jagielski wrote:
>> This is weird... Are you using the stuff at the ASF??
>> The format warning are way wrong as well:
>>
>> /** Keyspace hits */
>> apr_uint64_t keyspace_hits;
>> /** Keyspace misses */
>> apr_uint64_t keyspace_misses;
>>
>> keyspace_* are obviously not apr_uint32_ts
>
> They're uint32s in my local copy too, and the version on the server:
>
> http://svn.apache.org/viewvc/apr/apr-util/branches/1.6.x/include/apr_redis.h?view=markup#l420
>
> Same for the version in APR 2.0. Did we miss some alternate instructions for the new code?
>
> --Jacob
Re: mod_socache_redis
Posted by Jacob Champion <ch...@gmail.com>.
On 11/06/2016 05:32 AM, Jim Jagielski wrote:
> This is weird... Are you using the stuff at the ASF??
> The format warning are way wrong as well:
>
> /** Keyspace hits */
> apr_uint64_t keyspace_hits;
> /** Keyspace misses */
> apr_uint64_t keyspace_misses;
>
> keyspace_* are obviously not apr_uint32_ts
They're uint32s in my local copy too, and the version on the server:
http://svn.apache.org/viewvc/apr/apr-util/branches/1.6.x/include/apr_redis.h?view=markup#l420
Same for the version in APR 2.0. Did we miss some alternate instructions
for the new code?
--Jacob
Re: mod_socache_redis
Posted by Jim Jagielski <ji...@jaguNET.com>.
This is weird... Are you using the stuff at the ASF??
The format warning are way wrong as well:
/** Keyspace hits */
apr_uint64_t keyspace_hits;
/** Keyspace misses */
apr_uint64_t keyspace_misses;
keyspace_* are obviously not apr_uint32_ts
> On Nov 6, 2016, at 1:11 AM, Hank Ibell <hw...@gmail.com> wrote:
>
> Hello Jim,
>
> I am trying to build trunk, but I am getting a compiler error. It seems the structure apr_redis_stats_t is missing a rejected_connections field. I was able to compile by adding the missing field to apr_redis_stats_t of course.
>
> Here is the compiler output (built on macOS):
>
> mod_socache_redis.c:332:29: error: no member named 'rejected_connections' in 'apr_redis_stats_t'
> stats->rejected_connections);
> ~~~~~ ^
> mod_socache_redis.c:334:22: warning: format specifies type 'unsigned long' but the argument has type 'apr_uint32_t' (aka 'unsigned int') [-Wformat]
> stats->keyspace_hits, stats->keyspace_misses);
> ^~~~~~~~~~~~~~~~~~~~
> mod_socache_redis.c:334:44: warning: format specifies type 'unsigned long' but the argument has type 'apr_uint32_t' (aka 'unsigned int') [-Wformat]
> stats->keyspace_hits, stats->keyspace_misses);
> ^~~~~~~~~~~~~~~~~~~~~~
> mod_socache_redis.c:361:28: error: no member named 'rejected_connections' in 'apr_redis_stats_t'
> stats->rejected_connections, br);
> ~~~~~ ^
> mod_socache_redis.c:363:21: warning: format specifies type 'unsigned long' but the argument has type 'apr_uint32_t' (aka 'unsigned int') [-Wformat]
> stats->keyspace_hits, stats->keyspace_misses, br);
> ^~~~~~~~~~~~~~~~~~~~
> mod_socache_redis.c:363:43: warning: format specifies type 'unsigned long' but the argument has type 'apr_uint32_t' (aka 'unsigned int') [-Wformat]
> stats->keyspace_hits, stats->keyspace_misses, br);
> ^~~~~~~~~~~~~~~~~~~~~~
>
> On Sat, Nov 5, 2016 at 6:24 PM, Jim Jagielski <ji...@jagunet.com> wrote:
> Trunk now includes mod_socache_redis, which requires APR 1.6
> and above. Please test it out and provide feedback :)
>
> I also added in the required mod_status hook info for mod_socache_memcache
> as well, so now it was also provide mod_status info. I plan to
> also submit that for backporting to 2.4 as well ;)
>
Re: mod_socache_redis
Posted by Jim Jagielski <ji...@jaguNET.com>.
Is this Xcode8 or 7?
I'm on OSX 10.11.6 and Xcode 7.3.1
> On Nov 6, 2016, at 1:11 AM, Hank Ibell <hw...@gmail.com> wrote:
>
> Hello Jim,
>
> I am trying to build trunk, but I am getting a compiler error. It seems the structure apr_redis_stats_t is missing a rejected_connections field. I was able to compile by adding the missing field to apr_redis_stats_t of course.
>
> Here is the compiler output (built on macOS):
>
> mod_socache_redis.c:332:29: error: no member named 'rejected_connections' in 'apr_redis_stats_t'
> stats->rejected_connections);
> ~~~~~ ^
> mod_socache_redis.c:334:22: warning: format specifies type 'unsigned long' but the argument has type 'apr_uint32_t' (aka 'unsigned int') [-Wformat]
> stats->keyspace_hits, stats->keyspace_misses);
> ^~~~~~~~~~~~~~~~~~~~
> mod_socache_redis.c:334:44: warning: format specifies type 'unsigned long' but the argument has type 'apr_uint32_t' (aka 'unsigned int') [-Wformat]
> stats->keyspace_hits, stats->keyspace_misses);
> ^~~~~~~~~~~~~~~~~~~~~~
> mod_socache_redis.c:361:28: error: no member named 'rejected_connections' in 'apr_redis_stats_t'
> stats->rejected_connections, br);
> ~~~~~ ^
> mod_socache_redis.c:363:21: warning: format specifies type 'unsigned long' but the argument has type 'apr_uint32_t' (aka 'unsigned int') [-Wformat]
> stats->keyspace_hits, stats->keyspace_misses, br);
> ^~~~~~~~~~~~~~~~~~~~
> mod_socache_redis.c:363:43: warning: format specifies type 'unsigned long' but the argument has type 'apr_uint32_t' (aka 'unsigned int') [-Wformat]
> stats->keyspace_hits, stats->keyspace_misses, br);
> ^~~~~~~~~~~~~~~~~~~~~~
>
> On Sat, Nov 5, 2016 at 6:24 PM, Jim Jagielski <ji...@jagunet.com> wrote:
> Trunk now includes mod_socache_redis, which requires APR 1.6
> and above. Please test it out and provide feedback :)
>
> I also added in the required mod_status hook info for mod_socache_memcache
> as well, so now it was also provide mod_status info. I plan to
> also submit that for backporting to 2.4 as well ;)
>
Re: mod_socache_redis
Posted by Hank Ibell <hw...@gmail.com>.
Hello Jim,
I am trying to build trunk, but I am getting a compiler error. It seems the
structure apr_redis_stats_t is missing a rejected_connections field. I was
able to compile by adding the missing field to apr_redis_stats_t of course.
Here is the compiler output (built on macOS):
mod_socache_redis.c:332:29: error: no member named 'rejected_connections'
in 'apr_redis_stats_t'
stats->rejected_connections);
~~~~~ ^
mod_socache_redis.c:334:22: warning: format specifies type 'unsigned long'
but the argument has type 'apr_uint32_t' (aka 'unsigned int') [-Wformat]
stats->keyspace_hits, stats->keyspace_misses);
^~~~~~~~~~~~~~~~~~~~
mod_socache_redis.c:334:44: warning: format specifies type 'unsigned long'
but the argument has type 'apr_uint32_t' (aka 'unsigned int') [-Wformat]
stats->keyspace_hits, stats->keyspace_misses);
^~~~~~~~~~~~~~~~~~~~~~
mod_socache_redis.c:361:28: error: no member named 'rejected_connections'
in 'apr_redis_stats_t'
stats->rejected_connections, br);
~~~~~ ^
mod_socache_redis.c:363:21: warning: format specifies type 'unsigned long'
but the argument has type 'apr_uint32_t' (aka 'unsigned int') [-Wformat]
stats->keyspace_hits, stats->keyspace_misses, br);
^~~~~~~~~~~~~~~~~~~~
mod_socache_redis.c:363:43: warning: format specifies type 'unsigned long'
but the argument has type 'apr_uint32_t' (aka 'unsigned int') [-Wformat]
stats->keyspace_hits, stats->keyspace_misses, br);
^~~~~~~~~~~~~~~~~~~~~~
On Sat, Nov 5, 2016 at 6:24 PM, Jim Jagielski <ji...@jagunet.com> wrote:
> Trunk now includes mod_socache_redis, which requires APR 1.6
> and above. Please test it out and provide feedback :)
>
> I also added in the required mod_status hook info for mod_socache_memcache
> as well, so now it was also provide mod_status info. I plan to
> also submit that for backporting to 2.4 as well ;)
>