Thread: [Quickfix-developers] problem with invalid PossDup=Y messages
Brought to you by:
orenmnero
Menu
▾
▴
quickfix-developers
|
From: Hugo L. <Hug...@in...> - 2004-05-04 18:30:20
|
Hi all, I've detected a problem between quickfix server + client (or: acceptor + initiator). Let's suppose this rare scenario: - client logs on; - client detects seqnum gap, asks to resend messages from, say, seqnum=20; - server resends message 20; - (somehow message 20 has a tag without value, or some other validation problem); - client rejects message 20; (but does not increase targetseqnum) - eventually, server sends next message; - client detects seqnum gap, asks to resend messages from seqnum=20; - server resends message 20; - (etc, etc, client and server become stuck in this resend loop) Versions affected (possibly not limited to) : 1.3.2, 1.7.0. Am I doing something wrong / is this not true ? Thanks in advance, Hugo Leote |
|
From: Oren M. <or...@qu...> - 2004-05-04 18:40:47
|
Hugo, This has been reported several times in the past couple weeks. We are working on getting a fix for this into the next point release. --oren On May 4, 2004, at 1:29 PM, Hugo Leote wrote: > Hi all, > > I've detected a problem between quickfix server + client (or: acceptor > + > initiator). > Let's suppose this rare scenario: > > - client logs on; > - client detects seqnum gap, asks to resend messages from, say, > seqnum=20; > - server resends message 20; > - (somehow message 20 has a tag without value, or some other validation > problem); > - client rejects message 20; (but does not increase targetseqnum) > - eventually, server sends next message; > - client detects seqnum gap, asks to resend messages from seqnum=20; > - server resends message 20; > - (etc, etc, client and server become stuck in this resend loop) > > > Versions affected (possibly not limited to) : 1.3.2, 1.7.0. > > Am I doing something wrong / is this not true ? > > > Thanks in advance, > Hugo Leote > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: Oracle 10g > Get certified on the hottest thing ever to hit the market... Oracle > 10g. > Take an Oracle 10g class now, and we'll give you the exam FREE. > http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click > _______________________________________________ > Quickfix-developers mailing list > Qui...@li... > https://lists.sourceforge.net/lists/listinfo/quickfix-developers > |
|
From: Caleb E. <ca...@bk...> - 2004-05-04 18:45:40
|
On Tue, May 04, 2004 at 07:29:22PM +0100, Hugo Leote wrote:
> - client rejects message 20; (but does not increase targetseqnum)
This is a bug which I believe is being addressed in the next
release. I think it may already be fixed in CVS.
> - eventually, server sends next message;
> - client detects seqnum gap, asks to resend messages from seqnum=20;
> - server resends message 20;
> - (etc, etc, client and server become stuck in this resend loop)
>
> Versions affected (possibly not limited to) : 1.3.2, 1.7.0.
>
> Am I doing something wrong / is this not true ?
No, its a bug.
--
Caleb Epstein | bklyn . org | An age is called Dark not because the light
cae at | Brooklyn Dust | fails to shine, but because people refuse to
bklyn dot org | Bunny Mfg. | see it.
| | -- James Michener, "Space"
|
|
From: James C. D. <jc...@co...> - 2004-05-04 19:09:45
|
Hugo, We are seeing reports of this issue but I can't seem to replicate it with an acceptance test. Can you post the logs that show this issue? I'll keep trying to replicate it with a test. Jim James C. Downs Connamara Systems, LLC 53 W. Jackson Blvd Suite 1627 Chicago, IL 60604 312 - 282 - 7746 www.connamara.com -----Original Message----- From: qui...@li... [mailto:qui...@li...] On Behalf Of Hugo Leote Sent: Tuesday, May 04, 2004 1:29 PM To: qui...@li... Subject: [Quickfix-developers] problem with invalid PossDup=Y messages Hi all, I've detected a problem between quickfix server + client (or: acceptor + initiator). Let's suppose this rare scenario: - client logs on; - client detects seqnum gap, asks to resend messages from, say, seqnum=20; - server resends message 20; - (somehow message 20 has a tag without value, or some other validation problem); - client rejects message 20; (but does not increase targetseqnum) - eventually, server sends next message; - client detects seqnum gap, asks to resend messages from seqnum=20; - server resends message 20; - (etc, etc, client and server become stuck in this resend loop) Versions affected (possibly not limited to) : 1.3.2, 1.7.0. Am I doing something wrong / is this not true ? Thanks in advance, Hugo Leote ------------------------------------------------------- This SF.Net email is sponsored by: Oracle 10g Get certified on the hottest thing ever to hit the market... Oracle 10g. Take an Oracle 10g class now, and we'll give you the exam FREE. http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click _______________________________________________ Quickfix-developers mailing list Qui...@li... https://lists.sourceforge.net/lists/listinfo/quickfix-developers |
|
From: Oren M. <or...@qu...> - 2004-05-05 01:03:00
|
I checked in the tests and a fix for this. Jim, please verify that the test looks good. If anyone who has been having this problem can verify it goes when using the CVS version, please let me know. We can roll out a new release if all is good. On May 4, 2004, at 2:07 PM, James C. Downs wrote: > Hugo, > We are seeing reports of this issue but I can't seem to replicate it > with an > acceptance test. Can you post the logs that show this issue? I'll keep > trying to replicate it with a test. > > > Jim > > James C. Downs > Connamara Systems, LLC > 53 W. Jackson Blvd > Suite 1627 > Chicago, IL 60604 > 312 - 282 - 7746 > www.connamara.com > > -----Original Message----- > From: qui...@li... > [mailto:qui...@li...] On Behalf Of > Hugo > Leote > Sent: Tuesday, May 04, 2004 1:29 PM > To: qui...@li... > Subject: [Quickfix-developers] problem with invalid PossDup=Y messages > > Hi all, > > I've detected a problem between quickfix server + client (or: acceptor > + > initiator). > Let's suppose this rare scenario: > > - client logs on; > - client detects seqnum gap, asks to resend messages from, say, > seqnum=20; > - server resends message 20; > - (somehow message 20 has a tag without value, or some other validation > problem); > - client rejects message 20; (but does not increase targetseqnum) > - eventually, server sends next message; > - client detects seqnum gap, asks to resend messages from seqnum=20; > - server resends message 20; > - (etc, etc, client and server become stuck in this resend loop) > > > Versions affected (possibly not limited to) : 1.3.2, 1.7.0. > > Am I doing something wrong / is this not true ? > > > Thanks in advance, > Hugo Leote > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: Oracle 10g Get certified on the > hottest > thing ever to hit the market... Oracle 10g. > Take an Oracle 10g class now, and we'll give you the exam FREE. > http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click > _______________________________________________ > Quickfix-developers mailing list > Qui...@li... > https://lists.sourceforge.net/lists/listinfo/quickfix-developers > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: Oracle 10g > Get certified on the hottest thing ever to hit the market... Oracle > 10g. > Take an Oracle 10g class now, and we'll give you the exam FREE. > http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click > _______________________________________________ > Quickfix-developers mailing list > Qui...@li... > https://lists.sourceforge.net/lists/listinfo/quickfix-developers > |
|
From: James C. D. <jc...@co...> - 2004-05-05 11:40:02
|
Oren, I just checked out the latest version from CVS and all looks good. Thanks, Jim James C. Downs Connamara Systems, LLC 53 W. Jackson Blvd Suite 1627 Chicago, IL 60604 312 - 282 - 7746 www.connamara.com -----Original Message----- From: Oren Miller [mailto:or...@qu...] Sent: Tuesday, May 04, 2004 8:03 PM To: James C. Downs Cc: qui...@li...; 'Hugo Leote' Subject: Re: [Quickfix-developers] problem with invalid PossDup=Y messages I checked in the tests and a fix for this. Jim, please verify that the test looks good. If anyone who has been having this problem can verify it goes when using the CVS version, please let me know. We can roll out a new release if all is good. On May 4, 2004, at 2:07 PM, James C. Downs wrote: > Hugo, > We are seeing reports of this issue but I can't seem to replicate it > with an acceptance test. Can you post the logs that show this issue? > I'll keep trying to replicate it with a test. > > > Jim > > James C. Downs > Connamara Systems, LLC > 53 W. Jackson Blvd > Suite 1627 > Chicago, IL 60604 > 312 - 282 - 7746 > www.connamara.com > > -----Original Message----- > From: qui...@li... > [mailto:qui...@li...] On Behalf Of > Hugo Leote > Sent: Tuesday, May 04, 2004 1:29 PM > To: qui...@li... > Subject: [Quickfix-developers] problem with invalid PossDup=Y messages > > Hi all, > > I've detected a problem between quickfix server + client (or: acceptor > + > initiator). > Let's suppose this rare scenario: > > - client logs on; > - client detects seqnum gap, asks to resend messages from, say, > seqnum=20; > - server resends message 20; > - (somehow message 20 has a tag without value, or some other > validation problem); > - client rejects message 20; (but does not increase targetseqnum) > - eventually, server sends next message; > - client detects seqnum gap, asks to resend messages from seqnum=20; > - server resends message 20; > - (etc, etc, client and server become stuck in this resend loop) > > > Versions affected (possibly not limited to) : 1.3.2, 1.7.0. > > Am I doing something wrong / is this not true ? > > > Thanks in advance, > Hugo Leote > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: Oracle 10g Get certified on the > hottest thing ever to hit the market... Oracle 10g. > Take an Oracle 10g class now, and we'll give you the exam FREE. > http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click > _______________________________________________ > Quickfix-developers mailing list > Qui...@li... > https://lists.sourceforge.net/lists/listinfo/quickfix-developers > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: Oracle 10g > Get certified on the hottest thing ever to hit the market... Oracle > 10g. > Take an Oracle 10g class now, and we'll give you the exam FREE. > http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click > _______________________________________________ > Quickfix-developers mailing list > Qui...@li... > https://lists.sourceforge.net/lists/listinfo/quickfix-developers > |
|
From: Hugo L. <Hug...@in...> - 2004-05-05 11:16:25
|
Hello again, (sorry for the delay) I believe it's no longer necessary for me to post logs. I've just tested the CVS modifications on Session.cpp (rev 1.40) and the problem is solved ! Thank you Oren, Jim and Caleb for your feedback. Hugo Leote P.S.: The problem was solved in less than 10 hours. Can't beat that... -----Original Message----- From: "James C. Downs" <jc...@co...> To: "'Hugo Leote'" <Hug...@in...>, <quickfix- dev...@li...> Date: Tue, 4 May 2004 14:07:23 -0500 Subject: RE: [Quickfix-developers] problem with invalid PossDup=Y messages > Hugo, > We are seeing reports of this issue but I can't seem to replicate it > with an > acceptance test. Can you post the logs that show this issue? I'll keep > trying to replicate it with a test. > > > Jim > > James C. Downs > Connamara Systems, LLC > 53 W. Jackson Blvd > Suite 1627 > Chicago, IL 60604 > 312 - 282 - 7746 > www.connamara.com > > -----Original Message----- > From: qui...@li... > [mailto:qui...@li...] On Behalf Of > Hugo > Leote > Sent: Tuesday, May 04, 2004 1:29 PM > To: qui...@li... > Subject: [Quickfix-developers] problem with invalid PossDup=Y messages > > Hi all, > > I've detected a problem between quickfix server + client (or: acceptor > + > initiator). > Let's suppose this rare scenario: > > - client logs on; > - client detects seqnum gap, asks to resend messages from, say, > seqnum=20; > - server resends message 20; > - (somehow message 20 has a tag without value, or some other validation > problem); > - client rejects message 20; (but does not increase targetseqnum) > - eventually, server sends next message; > - client detects seqnum gap, asks to resend messages from seqnum=20; > - server resends message 20; > - (etc, etc, client and server become stuck in this resend loop) > > > Versions affected (possibly not limited to) : 1.3.2, 1.7.0. > > Am I doing something wrong / is this not true ? > > > Thanks in advance, > Hugo Leote > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: Oracle 10g Get certified on the > hottest > thing ever to hit the market... Oracle 10g. > Take an Oracle 10g class now, and we'll give you the exam FREE. > http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click > _______________________________________________ > Quickfix-developers mailing list > Qui...@li... > https://lists.sourceforge.net/lists/listinfo/quickfix-developers |
|
From: James C. D. <jc...@co...> - 2004-05-05 11:44:54
|
Hugo, Actually your description of the problem allowed Oren to create a test and fix the issue. Thanks for your help. Jim James C. Downs Connamara Systems, LLC 53 W. Jackson Blvd Suite 1627 Chicago, IL 60604 312 - 282 - 7746 www.connamara.com -----Original Message----- From: Hugo Leote [mailto:Hug...@in...] Sent: Wednesday, May 05, 2004 6:16 AM To: or...@qu...; James C. Downs; qui...@li... Subject: RE: [Quickfix-developers] problem with invalid PossDup=Y messages Hello again, (sorry for the delay) I believe it's no longer necessary for me to post logs. I've just tested the CVS modifications on Session.cpp (rev 1.40) and the problem is solved ! Thank you Oren, Jim and Caleb for your feedback. Hugo Leote P.S.: The problem was solved in less than 10 hours. Can't beat that... -----Original Message----- From: "James C. Downs" <jc...@co...> To: "'Hugo Leote'" <Hug...@in...>, <quickfix- dev...@li...> Date: Tue, 4 May 2004 14:07:23 -0500 Subject: RE: [Quickfix-developers] problem with invalid PossDup=Y messages > Hugo, > We are seeing reports of this issue but I can't seem to replicate it > with an acceptance test. Can you post the logs that show this issue? > I'll keep trying to replicate it with a test. > > > Jim > > James C. Downs > Connamara Systems, LLC > 53 W. Jackson Blvd > Suite 1627 > Chicago, IL 60604 > 312 - 282 - 7746 > www.connamara.com > > -----Original Message----- > From: qui...@li... > [mailto:qui...@li...] On Behalf Of > Hugo Leote > Sent: Tuesday, May 04, 2004 1:29 PM > To: qui...@li... > Subject: [Quickfix-developers] problem with invalid PossDup=Y messages > > Hi all, > > I've detected a problem between quickfix server + client (or: acceptor > + > initiator). > Let's suppose this rare scenario: > > - client logs on; > - client detects seqnum gap, asks to resend messages from, say, > seqnum=20; > - server resends message 20; > - (somehow message 20 has a tag without value, or some other > validation problem); > - client rejects message 20; (but does not increase targetseqnum) > - eventually, server sends next message; > - client detects seqnum gap, asks to resend messages from seqnum=20; > - server resends message 20; > - (etc, etc, client and server become stuck in this resend loop) > > > Versions affected (possibly not limited to) : 1.3.2, 1.7.0. > > Am I doing something wrong / is this not true ? > > > Thanks in advance, > Hugo Leote > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: Oracle 10g Get certified on the > hottest thing ever to hit the market... Oracle 10g. > Take an Oracle 10g class now, and we'll give you the exam FREE. > http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click > _______________________________________________ > Quickfix-developers mailing list > Qui...@li... > https://lists.sourceforge.net/lists/listinfo/quickfix-developers |
×
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.
Thanks for helping keep SourceForge clean.
X