[tickets] [opensaf:tickets] #2524 pyosaf: decorate function does not handle version struct in initialize functions

[Hieu Nguyen via Opensaf-tickets]

- **status**: unassigned --> accepted
- **assigned_to**: Hieu Nguyen
- **Comment**:

In progress



---

** [tickets:#2524] pyosaf: decorate function does not handle version struct in 
initialize functions**

**Status:** accepted
**Milestone:** 5.17.10
**Created:** Thu Jul 06, 2017 01:44 PM UTC by Zoran Milinkovic
**Last Updated:** Fri Jul 28, 2017 08:20 AM UTC
**Owner:** Hieu Nguyen


decorate() function does not handle version in initialize calls.
If initialize function returns ERR_TRY_AGAIN, version struct is populated with 
the latest version. The next initialize call will not initialize a handle with 
the required version, but with the latest library version


---

Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is 
subscribed to https://sourceforge.net/p/opensaf/tickets/

To unsubscribe from further messages, a project admin can change settings at 
https://sourceforge.net/p/opensaf/admin/tickets/options.  Or, if this is a 
mailing list, you can unsubscribe from the mailing list.--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Opensaf-tickets mailing list
Opensaf-tickets@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-tickets

[tickets] [opensaf:tickets] #2611 dtm: Allow the list of peer IP addresses to change in runtime

[Anders Widell via Opensaf-tickets]

---

** [tickets:#2611] dtm: Allow the list of peer IP addresses to change in 
runtime**

**Status:** assigned
**Milestone:** 5.17.10
**Created:** Wed Oct 04, 2017 11:55 AM UTC by Anders Widell
**Last Updated:** Wed Oct 04, 2017 11:55 AM UTC
**Owner:** Anders Widell


When using unicast, the list of peer IP addresses is retrieved from a file or 
by using a DNS query at startup. We can use e.g. inotify to detect when the 
file has been updated, and re-read the file when this happens. For the DNS 
case, we can make new DNS queries periodically.


---

Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is 
subscribed to https://sourceforge.net/p/opensaf/tickets/

To unsubscribe from further messages, a project admin can change settings at 
https://sourceforge.net/p/opensaf/admin/tickets/options.  Or, if this is a 
mailing list, you can unsubscribe from the mailing list.--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Opensaf-tickets mailing list
Opensaf-tickets@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-tickets

[tickets] [opensaf:tickets] #2610 dtm: Propagate peer IP addresses dynamically

[Anders Widell via Opensaf-tickets]

---

** [tickets:#2610] dtm: Propagate peer IP addresses dynamically**

**Status:** assigned
**Milestone:** 5.17.10
**Created:** Wed Oct 04, 2017 11:53 AM UTC by Anders Widell
**Last Updated:** Wed Oct 04, 2017 11:53 AM UTC
**Owner:** Anders Widell


When using unicast, the initial list of IP addresses may be become out of date 
when new peers are added. We can dynamically update the list of peer IP 
addresses using several mechanisms. The newly detected addresses should at 
least be stored in RAM inside osafdtmd, but can possibly also be written down 
to disk on persistent storage so that they are remembered after a restart of 
OpenSAF or even after a complete node reboot. The mechanism to detect new 
addresses include:

* Look at the source IP address of incoming TCP connections and check if it is 
a known or unknown (new) source
* Use some form of protocol (e.g. a gossip protocol) to exchange peer IP 
addresses between nodes that are connected with each other.

We probably also need some mechanism to purge the list of peers, when an 
auto-detected IP has not been in used for a period of time.


---

Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is 
subscribed to https://sourceforge.net/p/opensaf/tickets/

To unsubscribe from further messages, a project admin can change settings at 
https://sourceforge.net/p/opensaf/admin/tickets/options.  Or, if this is a 
mailing list, you can unsubscribe from the mailing list.--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Opensaf-tickets mailing list
Opensaf-tickets@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-tickets

[tickets] [opensaf:tickets] #2609 dtm: Handle unicast sender congestion

[Anders Widell via Opensaf-tickets]

---

** [tickets:#2609] dtm: Handle unicast sender congestion**

**Status:** unassigned
**Milestone:** 5.17.10
**Created:** Wed Oct 04, 2017 11:44 AM UTC by Anders Widell
**Last Updated:** Wed Oct 04, 2017 11:44 AM UTC
**Owner:** nobody


When using unicast instead of broadcast or multicast, we have the possibility 
to get congestion when sending the unicast messages to all peer nodes, 
resulting in success for some nodes and failure for other nodes. We should keep 
track of which nodes that have failed, and use e.g. EPOLLOUT to get notified 
when sending is possible again.


---

Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is 
subscribed to https://sourceforge.net/p/opensaf/tickets/

To unsubscribe from further messages, a project admin can change settings at 
https://sourceforge.net/p/opensaf/admin/tickets/options.  Or, if this is a 
mailing list, you can unsubscribe from the mailing list.--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Opensaf-tickets mailing list
Opensaf-tickets@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-tickets

[tickets] [opensaf:tickets] #2608 dtm: Add a configuration file for osaftransportd

[Anders Widell via Opensaf-tickets]

---

** [tickets:#2608] dtm: Add a configuration file for osaftransportd**

**Status:** unassigned
**Milestone:** 5.17.10
**Created:** Wed Oct 04, 2017 11:39 AM UTC by Anders Widell
**Last Updated:** Wed Oct 04, 2017 11:39 AM UTC
**Owner:** nobody


osaftransport currently doesn't have any configuration file in /etc/opensaf/. 
We should add one.


---

Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is 
subscribed to https://sourceforge.net/p/opensaf/tickets/

To unsubscribe from further messages, a project admin can change settings at 
https://sourceforge.net/p/opensaf/admin/tickets/options.  Or, if this is a 
mailing list, you can unsubscribe from the mailing list.--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Opensaf-tickets mailing list
Opensaf-tickets@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-tickets

[tickets] [opensaf:tickets] #2607 dtm: Separate CLC-CLI of osaftransportd and osafdtmd

[Anders Widell via Opensaf-tickets]

---

** [tickets:#2607] dtm: Separate CLC-CLI of osaftransportd and osafdtmd**

**Status:** unassigned
**Milestone:** 5.17.10
**Created:** Wed Oct 04, 2017 11:37 AM UTC by Anders Widell
**Last Updated:** Wed Oct 04, 2017 11:37 AM UTC
**Owner:** nobody


Currently, the opensaf-transport CLC-CLI script calls the opensaf-dtm CLC-CLI 
script. This mess should be cleaned up so that osaf-dtm is called directly from 
NID.


---

Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is 
subscribed to https://sourceforge.net/p/opensaf/tickets/

To unsubscribe from further messages, a project admin can change settings at 
https://sourceforge.net/p/opensaf/admin/tickets/options.  Or, if this is a 
mailing list, you can unsubscribe from the mailing list.--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Opensaf-tickets mailing list
Opensaf-tickets@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-tickets

[tickets] [opensaf:tickets] #2606 dtm: Remove configuration of TCP send and receive buffer size

[Anders Widell via Opensaf-tickets]

---

** [tickets:#2606] dtm: Remove configuration of TCP send and receive buffer 
size**

**Status:** unassigned
**Milestone:** 5.17.10
**Created:** Wed Oct 04, 2017 10:59 AM UTC by Anders Widell
**Last Updated:** Wed Oct 04, 2017 10:59 AM UTC
**Owner:** nobody


There is a configuration option DTM_SOCK_SND_RCV_BUF_SIZE for configuring the 
TCP send and receive buffer size. This option ought to be removed, since the 
kernel usually does a better job at setting the apropriate buffer size 
automatically.


---

Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is 
subscribed to https://sourceforge.net/p/opensaf/tickets/

To unsubscribe from further messages, a project admin can change settings at 
https://sourceforge.net/p/opensaf/admin/tickets/options.  Or, if this is a 
mailing list, you can unsubscribe from the mailing list.--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Opensaf-tickets mailing list
Opensaf-tickets@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-tickets

[tickets] [opensaf:tickets] #2121 log: parsing command logtest with option -e is not correct

[Srinivas Siva Mangipudy via Opensaf-tickets]

- **status**: review --> fixed



---

** [tickets:#2121] log: parsing command logtest with option -e is not correct**

**Status:** fixed
**Milestone:** future
**Created:** Mon Oct 17, 2016 06:56 AM UTC by Canh Truong
**Last Updated:** Wed Oct 04, 2017 10:06 AM UTC
**Owner:** Srinivas Siva Mangipudy


Parsing command logtest with option -e is wrong.
For example: Running command logtest -e 14 2 
All test cases in test suite 14 will be executed while we just expect 
that test case 2 of test suite 14 is executed.


---

Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is 
subscribed to https://sourceforge.net/p/opensaf/tickets/

To unsubscribe from further messages, a project admin can change settings at 
https://sourceforge.net/p/opensaf/admin/tickets/options.  Or, if this is a 
mailing list, you can unsubscribe from the mailing list.--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Opensaf-tickets mailing list
Opensaf-tickets@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-tickets

[tickets] [opensaf:tickets] #2121 log: parsing command logtest with option -e is not correct

[Srinivas Siva Mangipudy via Opensaf-tickets]

commit 8c725cd2d873aebd420e1857ece516ba8d0eab5b
Author: Srinivas 
Date:   Wed Oct 4 15:35:14 2017 +0530

log: corrected parsing of command logtest with option -e [#2121]

Instead of using strtok to parse the input args,
reading the args using index.




---

** [tickets:#2121] log: parsing command logtest with option -e is not correct**

**Status:** review
**Milestone:** future
**Created:** Mon Oct 17, 2016 06:56 AM UTC by Canh Truong
**Last Updated:** Thu Sep 21, 2017 10:46 AM UTC
**Owner:** Srinivas Siva Mangipudy


Parsing command logtest with option -e is wrong.
For example: Running command logtest -e 14 2 
All test cases in test suite 14 will be executed while we just expect 
that test case 2 of test suite 14 is executed.


---

Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is 
subscribed to https://sourceforge.net/p/opensaf/tickets/

To unsubscribe from further messages, a project admin can change settings at 
https://sourceforge.net/p/opensaf/admin/tickets/options.  Or, if this is a 
mailing list, you can unsubscribe from the mailing list.--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Opensaf-tickets mailing list
Opensaf-tickets@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-tickets

[tickets] [opensaf:tickets] #2546 log: debug assertion failed in log agent during scale-in

[Vu Minh Nguyen via Opensaf-tickets]

- **status**: not-reproducible --> fixed
- **Milestone**: never --> 5.17.10
- **Comment**:

Will re-open the ticket if encounter this fault again. Now consider it as fixed.



---

** [tickets:#2546] log: debug assertion failed in log agent during scale-in**

**Status:** fixed
**Milestone:** 5.17.10
**Created:** Tue Aug 08, 2017 07:56 AM UTC by Vu Minh Nguyen
**Last Updated:** Wed Oct 04, 2017 09:57 AM UTC
**Owner:** nobody


During scaling in, LOG application got coredump due to assertion failed in log 
agent:

> 320e :
> _ZN9LogClient17RestoreRefCounterE16RefCounterDegreeb():
> /src/log/agent/lga_client.cc:139 (discriminator 1)
> 320e: 48 8d 0d 8b a7 00 00lea0xa78b(%rip),%rcx# d9a0 
> 

Above disassembly points to this code line `assert(ref_counter_ >= -1);`

:::C++
void LogClient::RestoreRefCounter(RefCounterDegree value, bool updated) {
  TRACE_ENTER();
  if (updated == false) return;
  ScopeLock scopeLock(ref_counter_mutex_);
  ref_counter_ -= value;
  TRACE("%s: value = %d", __func__, ref_counter_);
  // Don't expect the @ref_counter_ is less than (-1)
  assert(ref_counter_ >= -1);
}




---

Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is 
subscribed to https://sourceforge.net/p/opensaf/tickets/

To unsubscribe from further messages, a project admin can change settings at 
https://sourceforge.net/p/opensaf/admin/tickets/options.  Or, if this is a 
mailing list, you can unsubscribe from the mailing list.--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Opensaf-tickets mailing list
Opensaf-tickets@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-tickets

[tickets] [opensaf:tickets] #2546 log: debug assertion failed in log agent during scale-in

[Vu Minh Nguyen via Opensaf-tickets]

- **status**: accepted --> not-reproducible
- **assigned_to**: Vu Minh Nguyen -->  nobody 
- **Milestone**: 5.17.10 --> never



---

** [tickets:#2546] log: debug assertion failed in log agent during scale-in**

**Status:** not-reproducible
**Milestone:** never
**Created:** Tue Aug 08, 2017 07:56 AM UTC by Vu Minh Nguyen
**Last Updated:** Tue Aug 29, 2017 02:27 AM UTC
**Owner:** nobody


During scaling in, LOG application got coredump due to assertion failed in log 
agent:

> 320e :
> _ZN9LogClient17RestoreRefCounterE16RefCounterDegreeb():
> /src/log/agent/lga_client.cc:139 (discriminator 1)
> 320e: 48 8d 0d 8b a7 00 00lea0xa78b(%rip),%rcx# d9a0 
> 

Above disassembly points to this code line `assert(ref_counter_ >= -1);`

:::C++
void LogClient::RestoreRefCounter(RefCounterDegree value, bool updated) {
  TRACE_ENTER();
  if (updated == false) return;
  ScopeLock scopeLock(ref_counter_mutex_);
  ref_counter_ -= value;
  TRACE("%s: value = %d", __func__, ref_counter_);
  // Don't expect the @ref_counter_ is less than (-1)
  assert(ref_counter_ >= -1);
}




---

Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is 
subscribed to https://sourceforge.net/p/opensaf/tickets/

To unsubscribe from further messages, a project admin can change settings at 
https://sourceforge.net/p/opensaf/admin/tickets/options.  Or, if this is a 
mailing list, you can unsubscribe from the mailing list.--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Opensaf-tickets mailing list
Opensaf-tickets@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-tickets

[tickets] [opensaf:tickets] #2039 NTF: Missleading returned code of ntftest when tests are failed

[Srinivas Siva Mangipudy via Opensaf-tickets]

- **status**: review --> fixed



---

** [tickets:#2039] NTF: Missleading returned code of ntftest when tests are 
failed**

**Status:** fixed
**Milestone:** 5.17.10
**Created:** Thu Sep 15, 2016 09:11 AM UTC by Minh Hon Chau
**Last Updated:** Wed Oct 04, 2017 09:30 AM UTC
**Owner:** Srinivas Siva Mangipudy


If running a test case of ntftest suite 37 (38, or 39) and it fails, the 
reported error is always SA_AIS_ERR_FAILED_OPERATION, but the actual failure 
code is different.
The actual error code should be memorized and reported in test result.


---

Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is 
subscribed to https://sourceforge.net/p/opensaf/tickets/

To unsubscribe from further messages, a project admin can change settings at 
https://sourceforge.net/p/opensaf/admin/tickets/options.  Or, if this is a 
mailing list, you can unsubscribe from the mailing list.--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Opensaf-tickets mailing list
Opensaf-tickets@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-tickets

[tickets] [opensaf:tickets] #2039 NTF: Missleading returned code of ntftest when tests are failed

[Srinivas Siva Mangipudy via Opensaf-tickets]

commit ac1d727d2b70cb41a612c2b9a377fccb355c8b19
Author: Srinivas 
Date:   Wed Oct 4 14:44:57 2017 +0530

ntf: correct return code is returned [#2039]



---

** [tickets:#2039] NTF: Missleading returned code of ntftest when tests are 
failed**

**Status:** review
**Milestone:** 5.17.10
**Created:** Thu Sep 15, 2016 09:11 AM UTC by Minh Hon Chau
**Last Updated:** Fri Sep 15, 2017 11:13 AM UTC
**Owner:** Srinivas Siva Mangipudy


If running a test case of ntftest suite 37 (38, or 39) and it fails, the 
reported error is always SA_AIS_ERR_FAILED_OPERATION, but the actual failure 
code is different.
The actual error code should be memorized and reported in test result.


---

Sent from sourceforge.net because opensaf-tickets@lists.sourceforge.net is 
subscribed to https://sourceforge.net/p/opensaf/tickets/

To unsubscribe from further messages, a project admin can change settings at 
https://sourceforge.net/p/opensaf/admin/tickets/options.  Or, if this is a 
mailing list, you can unsubscribe from the mailing list.--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Opensaf-tickets mailing list
Opensaf-tickets@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-tickets