The feature of having multiple certificates served on the same ip/port
is Server Name Indication (SNI) and the nsopenssl driver does not
support it. As you said, getting that to work would require some rewiring.
However, I think the certificate described by Thorpe was a single
certificate that
Brad Chick wrote:
> AOLserver community,
>
> I need to generate and parse json in my aolserver web app. Seems like it
> should be easy enough. But there are no aolserver modules that I can find.
>
> I did find: tcljson
> @
> https://code.google.com/p/aolserver/source/browse/trunk/modules/tcljson/?r
Sep Ng wrote:
> Thank you very much for shedding a lot of light into this.
>
> On Friday, March 20, 2015 at 3:58:19 PM UTC+8, Gustaf Neumann wrote:
>
> > Am 20.03.15 um 07:48 schrieb Sep Ng:
>
> what is hurting you?
>
> > We have instances where we'd get a high number of concurren
Brad Chick wrote:
> On the old AOLserver, under modules, there is nssession. But I can't
> seem to get it. cvs seems not to be working:
>
> [brad@repos ~]$ cvs -d :pserver:anonym...@cvs.panoptic.com:/home/cvs co
> -d nssession aolserver/nssession
> cvs [checkout aborted]: connect to [cvs.panoptic.c
Do you happen to have a custom error page for 503 errors (service
unavailable)? I'm looking over the code for places where a connection
might be dropped without sending anything, and although this path
doesn't seem to do that it does look possible for it to do other bad
things (i.e., crash).
Hi Cyan,
Yes, there's still a few subscribers.
Do you know if this connection dropping happens mostly when there is a
lot of activity or more frequently when there is very low activity?
I recall a few edge cases in the thread pooling where a thread would in
some circumstances wait until anothe
Ok, I've been sitting on this response for far too long now, I may as
well just send it out, perfection be damned.
That was probably my thread on "4.6 and beyond" you mentioned. Lots of
energy ... then life happened, and time vanished. I'd still love to
make things happen, I'm just short on e
Ian Harding wrote:
> I have a mountain of code in TCL, and I am thinking of rewriting some of
> it in C.
>
> In a C module I have a function that is registered to an URL. Like I
> said, I have a bunch of TCL code that I would like to be able to use
> until (if ever) it gets rewritten.
>
> I see tha
It depends on why the connection has been running a long time. The
"ns_loop_ctl" command lets you examing running loops (i.e., while and
foreach loops) and lets you interrupt them in case something got back a
much larger dataset than expected. If it's just a slow network
connection not lettin
It's not that hard to believe that a distro could have a link error in a
less-used package (which aolserver probably is, alas). Especially since
that particular missing symbol error is (if it is what I think it is)
fairly new and related to gcc changes in default symbol visibility, and
the fix
Wolfgang Winkler wrote:
> Hi!
>
> We have a couple of servers running as openvz guests and on bare metal.
> Some of them are much slower due to a high latency when calling even
> very small pages. The strace output (see below) suggests, that the
> interpreters are initialized on each call.
Are the
If this is the bug I think it is, a checkin from 10/2011 fixed this bug.
handle internal error from redirect recursion overflow
directly instead of redirecting to internal error page. Prevents
error displaying error page from crashing server.
If updating to a more recent server isn't an option,
Try using nspostgres off sourceforge - that appears to be the most
recent version.
Alternately, add nsdb to the libraries to link against in the makefile.
-J
Cesáreo García Rodicio wrote:
> Hi Jim
>
> I had already tried but ...
>
> > make AOLSERVER=/usr/local/aolserver ACS=1 POSTGRES=PG_CON
Maurizio Martignano wrote:
> 2.A set of necessary cosmetics/make up changes to the overall code base
> to make it more compliant with nowadays C STDs, and therefore more
> “acceptable” to nowadays C compilers, they are:
>
> a.I have made explicit all type conversions (with explicit casts)
>
> b.I
I think the decision to make incompatible api changes was a misstep in
naviserver. I understand the reasons - the arguments were useless,
there are better facilities to replace it, it cleans up the code a lot -
but I think compatibility is really important. The tcl core team places
an extraor
Wolfgang Winkler wrote:
> Hi!
>
> I've narrowed down my problem and it's definitely a problem with TLS.
I can get a crash with just tcl/threads/tls, so I would say that the
problem is not with aolserver.
sample code:
package require Thread
for {set x 0} {$x < 100} {incr x} {
thread
Frank Bergmann wrote:
> Hi Jeff,
>
>
> Perfect, thanks a lot, I pretty much agree with your view.
>
>> SPDY and WebSockets
>
> I may underestimate the importance of these protocols. I
> believe I haven't every heard about a real-world application
> using these protocols. I would be interested to he
Frank Bergmann wrote:
> Hi!
>
>
> I believe that it is important to consider the current
> phase of the "live cycle" of AOLserver in order to focus
> the roadmap on the important stuff:
>
> - AOLserver is "mature" software, there were relatively
>few changes in the last few years. There are ver
aT wrote:
> Feature request .
>
> - possibility to reload configuration, without restart.
Reloading the entire configuration would be difficult, as it could mean
re-initializing various data structures and loading libraries and
modules, which generally amounts to restarting the server anyway. I
Wolfgang Winkler wrote:
> Hi!
>
> I've cleaned up our websockets code a bit (ah, the joy of it), it should
> run now on a basic AOLserver install. Where should I post it?
If we had a contrib directory, you could put it there - but we don't
(mental note: add contrib directory).
If it's not too bi
Gustaf Neumann wrote:
> On 28.09.12 12:57, Wolfgang Winkler wrote:
>> Hi!
The only problem we have is similar to Johns occasional crashes. But
I have 2 test servers and 1 pre production server, where AOLserver
crashes with " Fatal: received fatal signal 11" or "alloc: invalid
bl
John Buckman wrote:
> SUGGESTIONS FOR A FUTURE AOLSERVER
>
> The "state of the art" is, I think, happening in the javascript world,
> with things such as node.js. If the aolserver community were really
> interested in getting new users, making it a top notch
> embedded-javascript web server would b
jgdavid...@mac.com wrote:
>
> and those NS_IMPORT/NS_EXPORT
> declarations.
Those are less of a windows-only thing now. Some recent version of gcc
introduced visibility attributes which are supposed to be super good for
improving link times or something like that, so unix with a new-enough
gcc
Maurizio Martignano wrote:
> Dear all,
> I do not think that removing Windows specific code is a good idea.
Hi Maurizio,
You make a good argument for keeping windows support. There is a cost
of some added complexity, but I don't think that's unreasonable.
The biggest challenges are keepi
jgdavid...@mac.com wrote:
>
> Hi,
>
> Every few years we talk about what's next for the strategic direction
> of AOLserver which is great. In addition to the ideas below (which
> are cool), I always bring up this question: Should we dump the
> Windows port in favor of a clean Unix code base, conf
Ayan George wrote:
> Can we:
>
> * Allow c99 language features.
This might impact platform support, or that might be a non-issue. Any
specifics on what you're looking for?
> * Fully support HTTP 1.1
I thought about this but without digging in I couldn't recall which
pieces are missing (other
Wolfgang Winkler wrote:
> Hi!
>
> Concerning websockets: I've written a tcl only solution for AOLserver
> which works very well for pushing data to thousands of clients. The only
> problem is the receiving end, as I've to poll all open sockets in
> [ns_chan -shared] because I don't get the events o
Sep Ng wrote:
> I say this half in jest but if tuning Aolserver can become less like
> black magic, that would be a huge plus!
Yeah, this is what I was thinking about with some better docs and a
configuration ui, maybe some auto-configuration scripts that look at
system memory size and the like.
Daniël Mantione wrote:
> IPV6, please, please, please!
>
> Seriously, lack of IPV6 support is the perfect excuse to phase out an
> AOLServer installation.
>
> Depending on how long it will take to produce 4.6, maybe even consider a
> 4.5 series release with ipv6, I have noticed any instabilities o
Hi all,
There should be a 4.5.2 final release sometime soon, but what comes
next? I've been organizing my wishlist of what I'd like to see in
future AOLserver releases and I'm throwing it out there for anyone else
to add to or comment on. These are not in any particular order; some
are half-
ns_sets are not internally interlocked. If you are using the same
shared set in multiple threads, you need to protect it with a mutex.
Do you specifically need the indexability of ns_sets? nsvs are easier to
use for most cases, and if you have more than a few keys probably faster
too.
-J
Sep Ng
Greetings All!
I've uploaded a new source bundle aolserver-4.5.2-rc0.tar.gz to the SF
files repository.
4.5.2 will be primarily a bugfix release, collecting various patches and
contributions since the previous release more than 3 years ago. Among
the important changes are:
* pre-queue filter
so this should have no immediate impact :)
-J
Jeff Rogers wrote:
> I'd guess this is related to the code to rebuild namespaces, there may
> be something different enough about tcloo namespaces to break it. It
> could be something as simple as using "namespace" instead of
Looks like that flag is being pulled in from the tcl compilation flags.
GCC claims it's a great thing, I suppose there's some rationale to it.
In any case, it can also be fixed by updating the definition of
NS_EXPORT in include/nsthread.h:
#else
# if __GNUC__ >= 4
# define NS_EXPORT
I'd guess this is related to the code to rebuild namespaces, there may
be something different enough about tcloo namespaces to break it. It
could be something as simple as using "namespace" instead of
"::namespace". I'll try building with the latest tcl and see what I run
into.
-J
John Buck
Abrantes Araújo Silva Filho wrote:
> The "ns_ora clob_dml_file" correctly inserted a small file as clob in
> the test table.
>
> The problem is with "ns_ora clob_get_file": when selecting the clob
> and writing it to a return file, the resulting file have a binary
> charset, and so a diff between t
Jeff Rogers wrote:
> I'm confusing myself now tho, because it certainly looks like the
> threads are only ever reaped by the driver thread, which should
> absolutely finish Ns_ThreadCreate before it can call it again. Conn
> threads create a new thread to replace themsel
Jim Davidson wrote:
>
> Howdy,
>
> The leak is more precisely bloat, i.e., Tcl interps that create more
> and more objects and [...] The memory is not
> actually "returned" via a munmap, e.g.. Nate Folkman and I spent
> tons of time trying to figure this out years ago with all sorts of
> coalesc
Bjoern Kiesbye wrote:
> Hi Thorpe, I think Aolserver 4.5.0 ignores this settings from
> config.tcl, and if not set explicitly using ns_pools set command,
> Aolserver will use deafult values. I would suggest to in crease the
> maxconns value to at least 100, it does not cost extra pre occupied
> re
First thought, do you have a firewall that might be blocking incoming
network traffic on that port?
Can you connect from a non-web browser (e.g., telnet or nc)? Maybe
chrome is doing something odd.
-J
Thorpe Mayes wrote:
> Hi,
>
> My guess is that this is simple, but it eludes me.
>
> I have
You can get a list of running threads with [ns_info threads], one of the
items returned there is the threadid which should correspond to a value
returned from [ns_thread id]
Alternately, you could have your detached threads set some richer state
info than running/not running in a nsv and use th
Yes, there is no way to remove a filter other than by restarting the
server. The best you can do without restarting is to redefine the
filter proc to be a no-op.
-J
Thorpe Mayes wrote:
> Hi,
>
> I have a simple (minded) question re ns_register_filter:
>
> How do I disable this?
>
> Remove the
Sep Ng wrote:
> Every once in a while on my aolserver logs, I find this error and I'm
> not sure what it means. Typically, I see it when performing a database
> operation and then aolserver drops into an error like:
> invalid database_id: nsdb0
This means that there is no handle associated with th
The default maximum length for a single line in the header is 4096
bytes, which your cookie will end up slightly longer than. An error
message should be logged in the server log, something line "max line
exceeded". It looks like the code just closes the socket when that
happens, tho I have a
Majid,
Can you come up with a code sample or use case that demonstrates the
buggy behavior? That's the best way to show that it actually is a
problem, and that the changes fix the problem.
-J
Majid Khan wrote:
> Hello All,
> Just checking the code of nspostgres driver (downloaded from git) an
Obvious question: did you load and configure nsopenssl in your server?
-J
Thorpe Mayes wrote:
> Hi,
>
> I am trying to create a listener for PayPal transactions.
>
> When I do this:
>
> set page [ns_httpspost $url $rqset $qsset]
>
> where url = a https url
> rqset is empty
> qsset is a set of par
Hey all,
I'm in the process of making some additions to ns_http (especially a
"waitall" subcommand that waits for any outstanding request to finish;
useful when you're making a lot of requests) and I got to wondering if
ns_http would be better as an external module (where it might be more
flex
Were you doing these two downloads at the same time over the same link?
If so, you're probably just seeing effects of the network, not
anything about the server itself.
If OTOH you're doing each download separately on an otherwise idle
connection and idle server, that would be very strange.
48 matches
Mail list logo
