# When list element deleted MOP\_DELETE called twice?

**URL:** <https://dmap-community.ductus.global/t/when-list-element-deleted-mop-delete-called-twice/1337>\
**Category:** Core Engine and APIs\
**Created:** [June 18, 2017, 1:06am UTC](https://dmap-community.ductus.global/t/when-list-element-deleted-mop-delete-called-twice/1337 "2017-06-18T01:06:59Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![bkannadassan](https://avatars.discourse-cdn.com/v4/letter/b/f6c823/32.png) [@bkannadassan](https://dmap-community.ductus.global/u/bkannadassan)\
**Post date:** [June 18, 2017, 1:07am UTC](https://dmap-community.ductus.global/t/when-list-element-deleted-mop-delete-called-twice/1337/1 "2017-06-18T01:07:00Z")

</div>

Hi All!

I have a list which has a list inside it and which further has a list inside it…

list A {  
list B {  
list C {  
}  
}  
}

When I delete list A element and I do diff\_iterate I am getting iterated twice for MOP\_DELETE for A. Issue here is in the second iteration I am not having the element as its already deleted on first MOP\_DELETE call. Please let me know if I am missing something ?.

rgds  
Balaji

---

<div class="post-metadata">

**Author:** ![cohult](https://yyz2.discourse-cdn.com/flex010/user_avatar/dmap-community.ductus.global/cohult/32/221_2.png) [@cohult](https://dmap-community.ductus.global/u/cohult)\
**Post date:** [June 18, 2017, 1:48pm UTC](https://dmap-community.ductus.global/t/when-list-element-deleted-mop-delete-called-twice/1337/2 "2017-06-18T13:48:53Z")

</div>

Does it really happen twice? Have you checked if you have two subscription points that’s triggered?

Quick demo:

```
$ cat test.yang
module test {
  namespace "http://tail-f.com/ns/example/test";
  prefix t;
  list A {
    key a;
    leaf a { type string; }
    leaf a2 { type string; }
    list B {
      key b;
      leaf b { type string; }
      leaf b2 { type string; }
      list C {
        key c;
        leaf c { type string; }
        leaf c2 { type string; }
      }
    }
  }

$ cat test.xml
<config xmlns="http://tail-f.com/ns/config/1.0">
  <A xmlns="http://tail-f.com/ns/example/root">
    <a>testA</a>
    <a2>testA2</a2>
    <B>
      <b>testB</b>
      <b2>testB2</b2>
      <C>
        <c>testC</c>
        <c2>testC2</c2>
      </C>
    </B>
  </A>
</config>

$ confd_load -d -d -m -l test.xml
TRACE Connected (maapi) to ConfD
starting user session ctxt=system user=system groups=[system]
TRACE MAAPI_START_USER_SESSION --> CONFD_OK
TRACE MAAPI_START_TRANS --> CONFD_OK
TRACE MAAPI_LOAD_CONFIG_FILE --> CONFD_OK
TRACE MAAPI_APPLY_TRANS --> CONFD_OK
TRACE MAAPI_END_USER_SESSION --> CONFD_OK

$ confd_cmd -d -d -c "mdel /A" 
mdel "/A"
TRACE Connected (maapi) to ConfD
TRACE MAAPI_START_USER_SESSION --> CONFD_OK
TRACE MAAPI_START_TRANS --> CONFD_OK
TRACE MAAPI_DELETE /A --> CONFD_OK
TRACE MAAPI_APPLY_TRANS --> CONFD_OK
TRACE MAAPI_STOP_TRANS --> CONFD_OK
TRACE MAAPI_END_USER_SESSION --> CONFD_OK

$ confd_cmd -d -d -c "subwait_iter / 1 2"
subwait_iter "/" "1" "2"
TRACE Connected (maapi) to ConfD
TRACE MAAPI_LOAD_ALL_NS
TRACE MAAPI_LOAD_HASH_DB
TRACE Connected (cdb) to ConfD
TRACE CDB_SUBSCRIBE / --> CONFD_OK
TRACE CDB_SUBSCRIBE_DONE --> CONFD_OK
SUBSCRIBED TO /
TRACE CDB_SUBSCRIPTION_EVENT --> 7
COMMIT
TRACE CDB_SUB_PROGRESS --> CONFD_OK
TRACE CDB_SUB_ITERATE 7
TRACE CDB_SUB_PROGRESS --> CONFD_OK
/ /A{testA} created
TRACE CDB_SUB_PROGRESS --> CONFD_OK
/ /A{testA}/a set ( -> testA)
TRACE CDB_SUB_PROGRESS --> CONFD_OK
/ /A{testA}/a2 set ( -> testA2)
TRACE CDB_SUB_PROGRESS --> CONFD_OK
/ /A{testA}/B{testB} created
TRACE CDB_SUB_PROGRESS --> CONFD_OK
/ /A{testA}/B{testB}/b set ( -> testB)
TRACE CDB_SUB_PROGRESS --> CONFD_OK
/ /A{testA}/B{testB}/C{testC} created
TRACE CDB_SUB_PROGRESS --> CONFD_OK
/ /A{testA}/B{testB}/C{testC}/c set ( -> testC)
TRACE CDB_SUB_PROGRESS --> CONFD_OK
/ /A{testA}/B{testB}/C{testC}/c2 set ( -> testC2)
TRACE CDB_SUB_PROGRESS --> CONFD_OK
/ /A{testA}/B{testB}/b2 set ( -> testB2)
TRACE CDB_SUB_PROGRESS --> CONFD_OK
TRACE CDB_SYNC_SUB CDB_DONE_PRIORITY --> CONFD_OK
DONE

TRACE CDB_SUBSCRIPTION_EVENT --> 7
COMMIT
TRACE CDB_SUB_PROGRESS --> CONFD_OK
TRACE CDB_SUB_ITERATE 7
TRACE CDB_SUB_PROGRESS --> CONFD_OK
/ /A{testA} deleted
TRACE CDB_SUB_PROGRESS --> CONFD_OK
TRACE CDB_SYNC_SUB CDB_DONE_PRIORITY --> CONFD_OK
DONE
```

---

<div class="post-metadata">

**Author:** ![bkannadassan](https://avatars.discourse-cdn.com/v4/letter/b/f6c823/32.png) [@bkannadassan](https://dmap-community.ductus.global/u/bkannadassan)\
**Post date:** [June 20, 2017, 4:43am UTC](https://dmap-community.ductus.global/t/when-list-element-deleted-mop-delete-called-twice/1337/3 "2017-06-20T04:43:49Z")

</div>

Thanks a lot , in code I could see we are registering twice. Removing the same worked fine…

---

<div class="post-metadata">

**Author:** ![aruns17](https://avatars.discourse-cdn.com/v4/letter/a/a183cd/32.png) [@aruns17](https://dmap-community.ductus.global/u/aruns17)\
**Post date:** [June 22, 2017, 8:06pm UTC](https://dmap-community.ductus.global/t/when-list-element-deleted-mop-delete-called-twice/1337/4 "2017-06-22T20:06:11Z")

</div>

> [@cohult](#):
>
> TRACE MAAPI\_STOP\_TRANS –\> CONFD\_OK

On the same topic.

if we have subscription for /A and /A/B  
— when a del on List A done, do we get MOP\_DELETE notification for each element in list B ?

---

<div class="post-metadata">

**Author:** ![cohult](https://yyz2.discourse-cdn.com/flex010/user_avatar/dmap-community.ductus.global/cohult/32/221_2.png) [@cohult](https://dmap-community.ductus.global/u/cohult)\
**Post date:** [June 22, 2017, 9:36pm UTC](https://dmap-community.ductus.global/t/when-list-element-deleted-mop-delete-called-twice/1337/5 "2017-06-22T21:36:43Z")

</div>

> [@aruns17](#):
>
> if we have subscription for /A and /A/B — when a del on List A done, do we get MOP\_DELETE notification for each element in list B ?

Terminal Window 1:  
$ confd\_load -d -d -m -l test.xml  
$ confd\_cmd -d -d -c “subwait\_iter /A 1 100”

Terminal Window 2:  
$ confd\_cmd -d -d -c “subwait\_iter /A/B 1 100”

Terminal Window 3:  
$ confd\_cmd -d -d -c “mdel /A”

Let us know the result when you tried the above 😉

---

<div class="post-metadata">

**Author:** ![aruns17](https://avatars.discourse-cdn.com/v4/letter/a/a183cd/32.png) [@aruns17](https://dmap-community.ductus.global/u/aruns17)\
**Post date:** [June 22, 2017, 10:55pm UTC](https://dmap-community.ductus.global/t/when-list-element-deleted-mop-delete-called-twice/1337/6 "2017-06-22T22:55:12Z")

</div>

I passed the ITER\_WANT\_ANCESTOR\_DELETE flag to the iter in list B subscription handler and I am able to get notificiations to the iter function.

---

<div class="post-metadata">

**Author:** ![cohult](https://yyz2.discourse-cdn.com/flex010/user_avatar/dmap-community.ductus.global/cohult/32/221_2.png) [@cohult](https://dmap-community.ductus.global/u/cohult)\
**Post date:** [June 23, 2017, 8:31am UTC](https://dmap-community.ductus.global/t/when-list-element-deleted-mop-delete-called-twice/1337/7 "2017-06-23T08:31:17Z")

</div>

Thanks for sharing. For other readers from the confd\_lib\_cdb(3) man page under cdb\_diff\_iterate():

> MOP\_DELETED  
> The list entry, presence container, or optional leaf given by kp has been deleted.  
> If the subscription was triggered because an ancestor was deleted, the iter() function will not called at all if the delete was above the subscription point. However if the flag ITER\_WANT\_ANCESTOR\_DELETE is passed to cdb\_diff\_iterate() then deletes that trigger a descendant subscription will also generate a call to iter(), and in this case kp will be the path that was actually deleted