<!-- Kamailio Project uses GitHub Issues only for bugs in the code or feature requests. Please use this template only for bug reports.
If you have questions about using Kamailio or related to its configuration file, ask on sr-users mailing list:
* http://lists.kamailio.org/cgi-bin/mailman/listinfo/sr-users
If you have questions about developing extensions to Kamailio or its existing C code, ask on sr-dev mailing list:
* http://lists.kamailio.org/cgi-bin/mailman/listinfo/sr-dev
Please try to fill this template as much as possible for any issue. It helps the developers to troubleshoot the issue.
If there is no content to be filled in a section, the entire section can be removed.
You can delete the comments from the template sections when filling.
You can delete next line and everything above before submitting (it is a comment). -->
### Description
ims_registrar_pcscf module create a process with rank 1 (PROC_SIPINIT) if reginfo flag is open..
This is wrong because it is written that PROC_SIPINIT rank should be given only to the first worker process.
``` #define PROC_SIPINIT 1 /**< First (special) SIP worker - some modules do special processing in this child, like loading db data */ ```
This leads to a problem. ims_usrloc_pcscf loads registration records from the db twice.
The codes are demonstrated below.
Similar old issue exists here for another module: https://github.com/kamailio/kamailio/issues/975
### Troubleshooting
#### Reproduction
<!-- If the issue can be reproduced, describe how it can be done. -->
#### Debugging Data
<!-- If you got a core dump, use gdb to extract troubleshooting data - full backtrace, local variables and the list of the code at the issue location.
gdb /path/to/kamailio /path/to/corefile bt full info locals list
If you are familiar with gdb, feel free to attach more of what you consider to be relevant. -->
``` (paste your debugging data here) ```
#### Log Messages
<!-- Check the syslog file and if there are relevant log messages printed by Kamailio, add them next, or attach to issue, or provide a link to download them (e.g., to a pastebin site). -->
Here are fork operations and their ranks:
``` [root@n5gc-ims-dev src]# cat /usr/src/erhan5.log | grep init_mod_child | grep ims_usrloc_pcscf 0(10662) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 0 rank -127: ims_usrloc_pcscf [main] 1(10671) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 1 rank 1: ims_usrloc_pcscf [udp receiver child=0 sock=10.10.12.101:5060 (172.30.65.101:5060)] 3(10673) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 3 rank 3: ims_usrloc_pcscf [udp receiver child=2 sock=10.10.12.101:5060 (172.30.65.101:5060)] 2(10672) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 2 rank 2: ims_usrloc_pcscf [udp receiver child=1 sock=10.10.12.101:5060 (172.30.65.101:5060)] 5(10675) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 5 rank -1: ims_usrloc_pcscf [slow timer] 0(10662) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 0 rank 0: ims_usrloc_pcscf [main] 6(10679) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 6 rank -1: ims_usrloc_pcscf [timer] 4(10674) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 4 rank 4: ims_usrloc_pcscf [udp receiver child=3 sock=10.10.12.101:5060 (172.30.65.101:5060)] 7(10680) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 7 rank -1: ims_usrloc_pcscf [secondary timer] 8(10686) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 8 rank 1: ims_usrloc_pcscf [RegInfo Event Processor] 9(10687) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 9 rank -1: ims_usrloc_pcscf [Dialog Clean Timer] 10(10688) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 10 rank -2: ims_usrloc_pcscf [ctl handler] 11(10689) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 11 rank 5: ims_usrloc_pcscf [tcp receiver (generic) child=0] 12(10694) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 12 rank 6: ims_usrloc_pcscf [tcp receiver (generic) child=1] 13(10695) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 13 rank 7: ims_usrloc_pcscf [tcp receiver (generic) child=2] 14(10697) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 14 rank 8: ims_usrloc_pcscf [tcp receiver (generic) child=3] 15(10699) DEBUG: <core> [core/sr_module.c:845]: init_mod_child(): idx 15 rank -4: ims_usrloc_pcscf [tcp main process]
```
#### SIP Traffic
<!-- If the issue is exposed by processing specific SIP messages, grab them with ngrep or save in a pcap file, then add them next, or attach to issue, or provide a link to download them (e.g., to a pastebin site). -->
``` (paste your sip traffic here) ```
### Possible Solutions
<!-- If you found a solution or workaround for the issue, describe it. Ideally, provide a pull request with a fix. -->
In following code ims_registrar_pcscf module create a process with rank 1 (PROC_SIPINIT) : fork_process(PROC_SIPINIT, "RegInfo Event Processor", 1);
```
static int child_init(int rank) {
LM_DBG("Initialization of module in child [%d] \n", rank); if ((subscribe_to_reginfo == 1) && (rank == PROC_MAIN)) { LM_DBG("Creating RegInfo Event Processor process\n"); int pid = fork_process(PROC_SIPINIT, "RegInfo Event Processor", 1); if (pid < 0) return -1; //error if (pid == 0) { if (cfg_child_init()) return -1; //error reginfo_event_process(); } }
if (rank == PROC_MAIN || rank == PROC_TCP_MAIN) return 0; if (rank == 1) { /* init stats */ //TODO if parameters are modified via cfg framework do i change them? //update_stat( max_expires_stat, default_registrar_cfg.max_expires ); update_stat( max_contacts_stat, default_registrar_cfg.max_contacts ); update_stat( default_expire_stat, default_registrar_cfg.default_expires ); }
/* don't do anything for main process and TCP manager process */ if (rank == PROC_MAIN || rank == PROC_TCP_MAIN) return 0;
return 0; } ```
Here is how ims_usrloc_pcscf loads records from db at its child init callback function:
``` if (_rank==PROC_SIPINIT && db_mode!=DB_ONLY) { // if cache is used, populate domains from DB for( ptr=root ; ptr ; ptr=ptr->next) { LM_DBG("Preloading domain %.*s\n", ptr->name.len, ptr->name.s); if (preload_udomain(ul_dbh, ptr->d) < 0) { LM_ERR("child(%d): failed to preload domain '%.*s'\n", _rank, ptr->name.len, ZSW(ptr->name.s)); return -1; } } } ```
### Additional Information
* **Kamailio Version** - output of `kamailio -v`
``` it exists at latest version on master branch. [here](https://github.com/kamailio/kamailio/blob/master/src/modules/ims_registrar_p...) ```
* **Operating System**:
<!-- Details about the operating system, the type: Linux (e.g.,: Debian 8.4, Ubuntu 16.04, CentOS 7.1, ...), MacOS, xBSD, Solaris, ...; Kernel details (output of `uname -a`) -->
``` CentOS 7.1 ```
Can you try to replace PROC_SIPINIT with PROC_RPC in the code above and see if all works as expected? If yes, then it can be fixed in the repo.
I used PROC_NOCHLDINIT and it worked fine.
PROC_RPC is for RPC type processes. "RegInfo Event Processor" is continuously reading reginfo event FIFO and process the reginfo messages coming with Notify messages. It is not RPC process.
#define PROC_RPC -2 /**< RPC type process */
..
Ok, thanks for testing with PROC_NOCHLDINIT. Just for reference, the change from issue #975 used PROC_XWORKER (770c914d3). Lets wait a bit more for other comments and then it can be fixed in the repo.
For the records, PROC_RPC is listed as RPC type process, but it's actually used for generic non-sip-worker processes that need to send sip traffic or access the resources that are enabled in child_init() by the other modules.
By using PROC_RPC does not mean the process will listen automatically for rpc commands and do only that.
Anyhow, if using PROC_NOCHLDINIT works as expected, then its ok to use it.
@ErhanOnur could you paste the change patch that you tested here, or just create a pull request for it?
This issue is stale because it has been open 6 weeks with no activity. Remove stale label or comment or this will be closed in 2 weeks.
Closed #2809 as not planned.