Re: time-out for R2T PDUs?
Julian Satran <[email protected]> Thu, 6 Nov 2008 10:31:02 -0500
| Newsgroups | gmane.ietf.ips |
|---|---|
| Message-ID | <OFFFDF794E.92D4D3FA-ON852574F9.0054E919-852574F9.00553D83@il.ibm.com> |
This is a multipart message in MIME format. --===============1176055621== Content-Type: multipart/alternative; boundary="=_alternative 00553BF7852574F9_=" This is a multipart message in MIME format. --=_alternative 00553BF7852574F9_= Content-Type: text/plain; charset="US-ASCII" The target may try to finish the command with a "check condition" or some connection recovery before resorting to the big hammer. Julo From: "Erez Zilber" <[email protected]> To: Julian Satran/Haifa/IBM@IBMIL Cc: "Deva Ranganathan" <[email protected]>, [email protected] Date: 06/11/2008 10:15 Subject: Re: [Ips] time-out for R2T PDUs? There could be a situation in which NOP-IN PDUs from the target are sent successfully to the initiator, but some bug causes the initiator not to send the data-out PDU. According to what you say, the target may decide that it can't wait anymore and logout, right? Thanks, Erez On Thu, Nov 6, 2008 at 3:09 PM, Julian Satran <[email protected]> wrote: > Correct - there is no standard time-out. Regards, Julo > > > From: "Deva Ranganathan" <[email protected]> > To: "Erez Zilber" <[email protected]> > Cc: [email protected] > Date: 06/11/2008 04:20 > Subject: Re: [Ips] timeout for R2T PDUs? > ________________________________ > > > > > On Thu, Nov 6, 2008 at 2:46 PM, Deva Ranganathan > <[email protected]> wrote: > Erez, > > I think that this s a target implementation issue wherein the implementation > can choose the timeout depending on the availability of resources and the > software architecture. A target also has a NOP-IN sent periodically and can > use this to cleanup the stale resources as well. > > thanks > > Deva > HCL Technologies. > > > > On Thu, Nov 6, 2008 at 12:54 PM, Erez Zilber <[email protected]> wrote: > Hi, > > I couldn't find an answer to the following question in iSCSI spec: > > During the execution of a WRITE command, an R2T PDU is sent from the > target to the initiator. What should happen if a bug in the initiator > causes it not to send the data-out PDU? Is there any timeout that will > allow the target to abort this command? > > Thanks, > Erez > _______________________________________________ > Ips mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/ips > > _______________________________________________ > Ips mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/ips > > > --=_alternative 00553BF7852574F9_= Content-Type: text/html; charset="US-ASCII" <font size=2 face="sans-serif">The target may try to finish the command with a "check condition" or some connection recovery before resorting to the big hammer.</font> <br> <br><font size=2 face="sans-serif">Julo</font> <br> <br> <br> <table width=100%> <tr valign=top> <td><font size=1 color=#5f5f5f face="sans-serif">From:</font> <td><font size=1 face="sans-serif">"Erez Zilber" <[email protected]></font> <tr valign=top> <td><font size=1 color=#5f5f5f face="sans-serif">To:</font> <td><font size=1 face="sans-serif">Julian Satran/Haifa/IBM@IBMIL</font> <tr> <td valign=top><font size=1 color=#5f5f5f face="sans-serif">Cc:</font> <td><font size=1 face="sans-serif">"Deva Ranganathan" <[email protected]>, [email protected]</font> <tr valign=top> <td><font size=1 color=#5f5f5f face="sans-serif">Date:</font> <td><font size=1 face="sans-serif">06/11/2008 10:15</font> <tr valign=top> <td><font size=1 color=#5f5f5f face="sans-serif">Subject:</font> <td><font size=1 face="sans-serif">Re: [Ips] time-out for R2T PDUs?</font></table> <br> <hr noshade> <br> <br> <br><tt><font size=2>There could be a situation in which NOP-IN PDUs from the target are<br> sent successfully to the initiator, but some bug causes the initiator<br> not to send the data-out PDU. According to what you say, the target<br> may decide that it can't wait anymore and logout, right?<br> <br> Thanks,<br> Erez<br> <br> On Thu, Nov 6, 2008 at 3:09 PM, Julian Satran <[email protected]> wrote:<br> > Correct - there is no standard time-out. Regards, Julo<br> ><br> ><br> > From: "Deva Ranganathan" <[email protected]><br> > To: "Erez Zilber" <[email protected]><br> > Cc: [email protected]<br> > Date: 06/11/2008 04:20<br> > Subject: Re: [Ips] timeout for R2T PDUs?<br> > ________________________________<br> ><br> ><br> ><br> ><br> > On Thu, Nov 6, 2008 at 2:46 PM, Deva Ranganathan<br> > <[email protected]> wrote:<br> > Erez,<br> ><br> > I think that this s a target implementation issue wherein the implementation<br> > can choose the timeout depending on the availability of resources and the<br> > software architecture. A target also has a NOP-IN sent periodically and can<br> > use this to cleanup the stale resources as well.<br> ><br> > thanks<br> ><br> > Deva<br> > HCL Technologies.<br> ><br> ><br> ><br> > On Thu, Nov 6, 2008 at 12:54 PM, Erez Zilber <[email protected]> wrote:<br> > Hi,<br> ><br> > I couldn't find an answer to the following question in iSCSI spec:<br> ><br> > During the execution of a WRITE command, an R2T PDU is sent from the<br> > target to the initiator. What should happen if a bug in the initiator<br> > causes it not to send the data-out PDU? Is there any timeout that will<br> > allow the target to abort this command?<br> ><br> > Thanks,<br> > Erez<br> > _______________________________________________<br> > Ips mailing list<br> > [email protected]<br> > </font></tt><a href=https://www.ietf.org/mailman/listinfo/ips><tt><font size=2>https://www.ietf.org/mailman/listinfo/ips</font></tt></a><tt><font size=2><br> ><br> > _______________________________________________<br> > Ips mailing list<br> > [email protected]<br> > </font></tt><a href=https://www.ietf.org/mailman/listinfo/ips><tt><font size=2>https://www.ietf.org/mailman/listinfo/ips</font></tt></a><tt><font size=2><br> ><br> ><br> ><br> </font></tt> <br> <br> --=_alternative 00553BF7852574F9_=-- --===============1176055621== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Ips mailing list [email protected] https://www.ietf.org/mailman/listinfo/ips --===============1176055621==--