[
https://issues.apache.org/jira/browse/TS-1169?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Leif Hedstrom updated TS-1169:
------------------------------
Fix Version/s: 3.1.4
> in enable-debug and pristine_host_hdr=0 mode, TS will crash when purge a
> cached object which is after TSCacheUrlSet
> -------------------------------------------------------------------------------------------------------------------
>
> Key: TS-1169
> URL: https://issues.apache.org/jira/browse/TS-1169
> Project: Traffic Server
> Issue Type: Bug
> Affects Versions: 3.1.3, 3.0.4
> Environment: configure --enable-debug
> Reporter: Conan Wang
> Fix For: 3.1.4
>
>
> {code}
> #6 0x0000000100d269af in _ink_assert (a=0x1003c6be0 "md5a == md5b ||
> t_state.txn_conf->maintain_pristine_host_hdr", f=0x1003c514e "HttpSM.cc",
> l=3921) at ink_assert.cc:44
> #7 0x0000000100123e59 in HttpSM::do_cache_delete_all_alts (this=0x109d41190,
> cont=0x0) at HttpSM.cc:3921
> {code}
> in HttpSM::do_cache_delete_all_alts debug code, if a object key is rewritten
> by TSCacheUrlSet, md5a will not equal md5b, and it will crash because
> maintain_pristine_host_hdr = 0 ( when disable pristine_host_hdr ).
> Anyway, the cached object is purged successfully. Maybe it's just a wrong
> assert which didn't consider TSCacheUrlSet case.
> {code}
> #ifdef DEBUG
> INK_MD5 md5a;
> INK_MD5 md5b;
> t_state.hdr_info.client_request.url_get()->MD5_get(&md5a);
> t_state.cache_info.lookup_url->MD5_get(&md5b);
> ink_assert(md5a == md5b || t_state.txn_conf->maintain_pristine_host_hdr);
> #endif
> {code}
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
