httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ruediger Pluem <>
Subject Re: [PATCH] PR37145: data loss with httpd-2.0.55 reverse proxy method=post
Date Fri, 21 Oct 2005 21:46:44 GMT

On 10/21/2005 07:43 PM, William A. Rowe, Jr. wrote:
> Ruediger Pluem wrote:

> We might be better off using this fix (and documenting the usage of all get
> brigade calls w.r.t. transient buckets), while in 2.0.x we might want to
> return an allocated bucket in mod_ssl to ensure third party 2.0 modules,
> with
> this very same mis-assumptions, don't trip over this effect.  Thoughts?

Just to avoid confusion about which patch we are talking. For trunk/2.2.x I will
commit the patch that exchanges APR_BRIGADE_CONCAT with ap_save_brigade.

If I understand you correctly you think that it might make more sense to use

Index: modules/ssl/ssl_engine_io.c
--- modules/ssl/ssl_engine_io.c»(Revision 326481)
+++ modules/ssl/ssl_engine_io.c»(Arbeitskopie)
@@ -1324,7 +1324,7 @@
     /* Create a transient bucket out of the decrypted data. */
     if (len > 0) {
         apr_bucket *bucket =
-            apr_bucket_transient_create(inctx->buffer, len, f->c->bucket_alloc);
+            apr_bucket_heap_create(inctx->buffer, len, NULL, f->c->bucket_alloc);
         APR_BRIGADE_INSERT_TAIL(bb, bucket);

in the 2.0.x case to avoid third party modules falling in the same pit?

To be honest I would be -0.5 on that because

- The sideefects of aboves patch are currently unclear to me and we only apply it to
  the stable branch without having it in the trunk.

- As I stated, I think we face a typical setaside situation here. So it would be the
  duty of the third party module to fix this.

Nevertheless I agree that some third party modules might not give too much on how they *should*



View raw message