From lqin@sce.carleton.ca Wed Mar 3 20:36:14 2004 From: lqin@sce.carleton.ca (Liang Qin) Date: Wed, 03 Mar 2004 15:36:14 -0500 Subject: [Xorp-hackers] Q: Can't validate start of process protoC Message-ID: <404641BE.1080202@sce.carleton.ca> Hi XORP users, I'm porting a routing protocol "protoC" to XORP v.0.5, I have a problem when I launch this module in XORP. Here is part of error message: -------------------- New module has PID 4406 step1_done (protoC) step2 (protoC) validate XRL: >finder://protoC/common/0.1/get_status< [ 2004/03/03 15:23:35 WARNING xorp_rtrmgr:4403 XrlFinderTarget +376 finder_base.cc handle_finder_0_2_resolve_xrl ] Handling method for finder/0.2/resolve_xrl failed: XrlCmdError 102 Command failed Target "protoC" does not exist or is not enabled. *-*-*- start running protoC target ***************the protoC status is 3 -------------------------- This error message will repeat until this module is killed. The printout messages are added by me, we can see that the target is running from its status code. It'll work if I add a infinite loop late, like while(true) e.run(); e is an instance of EventLoop My question is 1. Does this error message mean that protoC process is not up? or rtrmgr cannot find the process? 2. Do I have to use the infinite loop? Thanks! L. Qin From orion@icir.org Wed Mar 3 22:36:09 2004 From: orion@icir.org (Orion Hodson) Date: Wed, 03 Mar 2004 14:36:09 -0800 Subject: [Xorp-hackers] Q: Can't validate start of process protoC In-Reply-To: Your message of "Wed, 03 Mar 2004 15:36:14 EST." <404641BE.1080202@sce.carleton.ca> Message-ID: <200403032236.i23Ma9E9087638@puma.icir.org> /-- Liang Qin wrote: | Hi XORP users, | | I'm porting a routing protocol "protoC" to XORP v.0.5, I have a problem | when I launch | this module in XORP. | | Here is part of error message: | -------------------- | New module has PID 4406 | step1_done (protoC) | step2 (protoC) | validate | XRL: >finder://protoC/common/0.1/get_status< | [ 2004/03/03 15:23:35 WARNING xorp_rtrmgr:4403 XrlFinderTarget +376 | finder_base.cc handle_finder_0_2_resolve_xrl ] Handling method for | finder/0.2/resolve_xrl failed: XrlCmdError 102 Command failed Target | "protoC" does not exist or is not enabled. | *-*-*- start running protoC target | ***************the protoC status is 3 | -------------------------- | This error message will repeat until this module is killed. | | The printout messages are added by me, we can see that the target is | running from its status code. | | It'll work if I add a infinite loop late, like | while(true) | e.run(); | | e is an instance of EventLoop | | My question is | 1. Does this error message mean that protoC process is not up? or rtrmgr | cannot find the process? Yes. | 2. Do I have to use the infinite loop? Yes. XORP is single-thread. All processing is done on the basis of timers and file descriptor. Events are dispatched out of the eventloop. Ideally, your application will just loop about eventloop.run() after start up and call it from just one place. A good example of this is the standalone Finder process in libxipc/finder_main.cc. For the time being, it's probably better to work with the current CVS image of XORP rather than the 0.5 source code. Several things have improved, including some Router Manager diagnostics have improved since then. Kind Regards - Orion From lqin@sce.carleton.ca Thu Mar 4 15:00:06 2004 From: lqin@sce.carleton.ca (Liang Qin) Date: Thu, 04 Mar 2004 10:00:06 -0500 Subject: [Xorp-hackers] Q: Can't validate start of process protoC In-Reply-To: <200403032236.i23Ma9E9087638@puma.icir.org> References: <200403032236.i23Ma9E9087638@puma.icir.org> Message-ID: <40474476.4010500@sce.carleton.ca> Orion Hodson wrote: >/-- Liang Qin wrote: >| Hi XORP users, >| >| I'm porting a routing protocol "protoC" to XORP v.0.5, I have a problem >| when I launch >| this module in XORP. >| >| Here is part of error message: >| -------------------- >| New module has PID 4406 >| step1_done (protoC) >| step2 (protoC) >| validate >| XRL: >finder://protoC/common/0.1/get_status< >| [ 2004/03/03 15:23:35 WARNING xorp_rtrmgr:4403 XrlFinderTarget +376 >| finder_base.cc handle_finder_0_2_resolve_xrl ] Handling method for >| finder/0.2/resolve_xrl failed: XrlCmdError 102 Command failed Target >| "protoC" does not exist or is not enabled. >| *-*-*- start running protoC target >| ***************the protoC status is 3 >| -------------------------- >| This error message will repeat until this module is killed. >| >| The printout messages are added by me, we can see that the target is >| running from its status code. >| >| It'll work if I add a infinite loop late, like >| while(true) >| e.run(); >| >| e is an instance of EventLoop >| >| My question is >| 1. Does this error message mean that protoC process is not up? or rtrmgr >| cannot find the process? > >Yes. > > But I checked the protoC module (target) status, it's in "PROC_READY". >| 2. Do I have to use the infinite loop? > >Yes. XORP is single-thread. All processing is done on the basis of timers >and file descriptor. Events are dispatched out of the eventloop. Ideally, >your application will just loop about eventloop.run() after start up and call >it from just one place. A good example of this is the standalone Finder >process in libxipc/finder_main.cc. > > > Yes, I fond the example in finer_main.cc, which uses infinite loop: while (gbl_sig_exit == false) { e.run(); } The problem is that there is another infinite loop in my protoC after launching the taget. I tried to put e.run() in this loop and remove the first one, it doesn't work. Thanks! Liang From lqin@sce.carleton.ca Thu Mar 11 16:45:20 2004 From: lqin@sce.carleton.ca (Liang Qin) Date: Thu, 11 Mar 2004 11:45:20 -0500 Subject: [Xorp-hackers] cannot add to RIB Message-ID: <405097A0.2050400@sce.carleton.ca> Hi XORP users, I use xorp 0.5 and now port a routing protocol X with IPv6 support to XORP. I can add table to RIB by send-add_igp_table6(..), and get callback message. But when X module tries to send a route to RIB by ribClient.send_add_route6(.....) always got : process was killed with signal 11 Module abnormally killed: X I trace down the code, and find out the problem is calling this function in rib_xif.cc _sender->send(x, callback(this, ......, ::unmarshall_add_route6, cb); it seems that unmarshall_add_route6() hasn't been executed. Could anyone tell me where the problem is? TIA Liang Qin From pavlin@icir.org Thu Mar 11 21:22:21 2004 From: pavlin@icir.org (Pavlin Radoslavov) Date: Thu, 11 Mar 2004 13:22:21 -0800 Subject: [Xorp-hackers] cannot add to RIB In-Reply-To: Message from Liang Qin of "Thu, 11 Mar 2004 11:45:20 EST." <405097A0.2050400@sce.carleton.ca> Message-ID: <200403112122.i2BLML8N006473@possum.icir.org> > Hi XORP users, > > I use xorp 0.5 and now port a routing protocol X with IPv6 support to XORP. > > I can add table to RIB by send-add_igp_table6(..), and get callback message. > But when X module tries to send a route to RIB by > ribClient.send_add_route6(.....) > always got : > process was killed with signal 11 > Module abnormally killed: X > > I trace down the code, and find out the problem is calling this function > in rib_xif.cc > > _sender->send(x, callback(this, ......, ::unmarshall_add_route6, cb); > > it seems that unmarshall_add_route6() hasn't been executed. > > Could anyone tell me where the problem is? Liang, Can you try to use the lastest XORP code from the CVS repository. It contains numerous fixes since xorp-0.5 release. For example, it contains a new StaticRoutes module (xorp/static_routes) which uses send_add_route6() to install static IPv6 routes in RIB->FEA->kernel. I just tested it again and it seems to work. Regards, Pavlin From lqin@sce.carleton.ca Fri Mar 12 20:22:20 2004 From: lqin@sce.carleton.ca (Liang Qin) Date: Fri, 12 Mar 2004 15:22:20 -0500 Subject: [Xorp-hackers] cannot add to RIB In-Reply-To: <200403112122.i2BLML8N006473@possum.icir.org> References: <200403112122.i2BLML8N006473@possum.icir.org> Message-ID: <40521BFC.9020000@sce.carleton.ca> Hi Pavlin, I'm compiling the code I downloaded from CVS. My question is StaticRoutes module? Is there any example in XORP which uses it? Thanks! Liang Pavlin Radoslavov wrote: >>Hi XORP users, >> >>I use xorp 0.5 and now port a routing protocol X with IPv6 support to XORP. >> >>I can add table to RIB by send-add_igp_table6(..), and get callback message. >>But when X module tries to send a route to RIB by >>ribClient.send_add_route6(.....) >>always got : >>process was killed with signal 11 >>Module abnormally killed: X >> >>I trace down the code, and find out the problem is calling this function >>in rib_xif.cc >> >>_sender->send(x, callback(this, ......, ::unmarshall_add_route6, cb); >> >>it seems that unmarshall_add_route6() hasn't been executed. >> >>Could anyone tell me where the problem is? >> >> > >Liang, > >Can you try to use the lastest XORP code from the CVS repository. It >contains numerous fixes since xorp-0.5 release. For example, it >contains a new StaticRoutes module (xorp/static_routes) which uses >send_add_route6() to install static IPv6 routes in RIB->FEA->kernel. >I just tested it again and it seems to work. > >Regards, >Pavlin > > > From pavlin@icir.org Fri Mar 12 20:29:25 2004 From: pavlin@icir.org (Pavlin Radoslavov) Date: Fri, 12 Mar 2004 12:29:25 -0800 Subject: [Xorp-hackers] cannot add to RIB In-Reply-To: Message from Liang Qin of "Fri, 12 Mar 2004 15:22:20 EST." <40521BFC.9020000@sce.carleton.ca> Message-ID: <200403122029.i2CKTP8N053670@possum.icir.org> > My question is StaticRoutes module? Is there any example in XORP > which uses it? StaticRoutes is like a (very dummy) routing protocol: it is configured with the static routes, and then it injects them into the RIB. You can find a sample configuration example inside rtrmgr/config.boot : protocols { static_routes { route4 10.10.0.0/16 { nexthop: 192.150.187.108 } route4 10.20.0.0/16 { nexthop: 192.150.187.108 metric: 10 } } } If you need to configure static IPv6 routes, then just add "route6" entries (with IPv6 addresses of course). Regards, Pavlin From lqin@sce.carleton.ca Fri Mar 12 20:43:27 2004 From: lqin@sce.carleton.ca (Liang Qin) Date: Fri, 12 Mar 2004 15:43:27 -0500 Subject: [Xorp-hackers] cannot add to RIB In-Reply-To: <200403122029.i2CKTP8N053670@possum.icir.org> References: <200403122029.i2CKTP8N053670@possum.icir.org> Message-ID: <405220EF.8060708@sce.carleton.ca> Thanks for your quick reply... The internal test is failed (1 of 18 tests failed) Fatal test_finder_events: 21704 FINDER +331 test_find_events.cc assert_ observer_watching ].......... The Linux kernel is 2.4.20-13.9.HOSTAP Should I install some RPMS? So StaticRoutes is a another routing module, and I still have to send route directly to RIB. But you did test the send_add_route6() and it works, didn't you? have a good weekend! Liang Pavlin Radoslavov wrote: >>My question is StaticRoutes module? Is there any example in XORP >>which uses it? >> >> > >StaticRoutes is like a (very dummy) routing protocol: it is >configured with the static routes, and then it injects them into the >RIB. You can find a sample configuration example inside >rtrmgr/config.boot : > >protocols { > static_routes { > route4 10.10.0.0/16 { > nexthop: 192.150.187.108 > } > route4 10.20.0.0/16 { > nexthop: 192.150.187.108 > metric: 10 > } > } >} > >If you need to configure static IPv6 routes, then just add "route6" >entries (with IPv6 addresses of course). > >Regards, >Pavlin > > > From pavlin@icir.org Fri Mar 12 21:45:19 2004 From: pavlin@icir.org (Pavlin Radoslavov) Date: Fri, 12 Mar 2004 13:45:19 -0800 Subject: [Xorp-hackers] cannot add to RIB In-Reply-To: Message from Liang Qin of "Fri, 12 Mar 2004 15:43:27 EST." <405220EF.8060708@sce.carleton.ca> Message-ID: <200403122145.i2CLjJ8N054155@possum.icir.org> > Thanks for your quick reply... > The internal test is failed (1 of 18 tests failed) > Fatal test_finder_events: 21704 FINDER +331 test_find_events.cc assert_ > observer_watching ].......... > > The Linux kernel is > 2.4.20-13.9.HOSTAP Hmmm, interesting. We have RedHat-7.3 with 2.4.20-28.7smp kernel. We are running automated tests every few hours on that machine and we haven't seen that error. Was there any other xorp process running while you were running the above test (e.g., maybe some leftovers, etc)? > Should I install some RPMS? Maybe no. In general, to run some of the internal tests you must have "bash" installed. Also, depends on your Linux distribution, you may have to install openssl-devel and ncurses (see the Linux entry in the BUILD_NOTES file), but if the compilation succeeded for you then you probably don't need them. > So StaticRoutes is a another routing module, and I still have to send > route directly to RIB. But you did test the send_add_route6() and it works, > didn't you? Sorry, I wasn't clear in my email about the purpose of StaticRoutes. Yes, StaticRoutes is just another routing module which happens to use send_add_route6(). So you can look in its source code to find-out whether you were doing something different in respect of registering your module with the RIB and sending routes to it. The whole StaticRoutes module is rather simple so I guess it will be relatively easy to find the relevant stuff, but let me know if you have any questions. Regards, Pavlin From lqin@sce.carleton.ca Mon Mar 15 19:22:45 2004 From: lqin@sce.carleton.ca (Liang Qin) Date: Mon, 15 Mar 2004 14:22:45 -0500 Subject: [Xorp-hackers] cannot add to RIB In-Reply-To: <200403122145.i2CLjJ8N054155@possum.icir.org> References: <200403122145.i2CLjJ8N054155@possum.icir.org> Message-ID: <40560285.6030400@sce.carleton.ca> Hi Pavli, Thanks! I installed and test the latest code on another Linux box (kernel 2.4.20-13.9), it seems OK because it finished "gmake check" process though there were some error messages appeared. I have some questions to ask: 1. Now the module x is killed again with more detailed info: FATAL xorp_x: xxx LIBXORP +37 eventloop.cc EventLoop ] Asserttion (instance_count ==1) failed ... process was killed with signal 6 Module abnormally killed:X I checked the Eventloop code, the error messages means that "the EventLoop::run has not been called". But I did set it run. 2. What dos IfMgrXrlMirror object do in the StaticRoutes code? Do I have to use it if the module only sends route to RIB? Thanks! Liang Pavlin Radoslavov wrote: >>Thanks for your quick reply... >>The internal test is failed (1 of 18 tests failed) >>Fatal test_finder_events: 21704 FINDER +331 test_find_events.cc assert_ >>observer_watching ].......... >> >>The Linux kernel is >>2.4.20-13.9.HOSTAP >> >> > >Hmmm, interesting. We have RedHat-7.3 with 2.4.20-28.7smp kernel. >We are running automated tests every few hours on that machine and >we haven't seen that error. Was there any other xorp process running >while you were running the above test (e.g., maybe some leftovers, >etc)? > > > >>Should I install some RPMS? >> >> > >Maybe no. >In general, to run some of the internal tests you must have "bash" >installed. Also, depends on your Linux distribution, you may have to >install openssl-devel and ncurses (see the Linux entry in the >BUILD_NOTES file), but if the compilation succeeded for you then you >probably don't need them. > > > >>So StaticRoutes is a another routing module, and I still have to send >>route directly to RIB. But you did test the send_add_route6() and it works, >>didn't you? >> >> > >Sorry, I wasn't clear in my email about the purpose of StaticRoutes. >Yes, StaticRoutes is just another routing module which happens to >use send_add_route6(). So you can look in its source code to >find-out whether you were doing something different in respect of >registering your module with the RIB and sending routes to it. >The whole StaticRoutes module is rather simple so I guess it will be >relatively easy to find the relevant stuff, but let me know if you >have any questions. > >Regards, >Pavlin >_______________________________________________ >Xorp-hackers mailing list >Xorp-hackers@icir.org >http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/xorp-hackers > > > From pavlin@icir.org Mon Mar 15 20:23:42 2004 From: pavlin@icir.org (Pavlin Radoslavov) Date: Mon, 15 Mar 2004 12:23:42 -0800 Subject: [Xorp-hackers] cannot add to RIB In-Reply-To: Message from Liang Qin of "Mon, 15 Mar 2004 14:22:45 EST." <40560285.6030400@sce.carleton.ca> Message-ID: <200403152023.i2FKNgRj016391@possum.icir.org> > Thanks! I installed and test the latest code on another Linux > box (kernel 2.4.20-13.9), it seems OK because it finished "gmake > check" process though there were some error messages appeared. Can you send those error messages. > I have some questions to ask: > 1. Now the module x is killed again with more detailed info: > FATAL xorp_x: xxx LIBXORP +37 eventloop.cc EventLoop ] Asserttion > (instance_count ==1) > failed > > ... > process was killed with signal 6 > Module abnormally killed:X > > I checked the Eventloop code, the error messages means that "the > EventLoop::run > has not been called". But I did set it run. The "EventLoop::run has not been called" comment applies to variable "last_ev_run". It looks like that you have tried to create more than one EventLoop instance within the same binary. Currently, for many reasons we don't support more than EventLoop instance. You would have to create a single instance, and then pass it as reference elsewhere. > 2. What dos IfMgrXrlMirror object do in the StaticRoutes code? Do I have > to use > it if the module only sends route to RIB? Currently it is not used (but will be used in the future). Its purpose is to track the status of the network interfaces. Thus if, say, a network interface goes down, then StaticRoutes will withdraw all static routes that use that interface. Regards, Pavlin From lqin@sce.carleton.ca Tue Mar 16 16:07:06 2004 From: lqin@sce.carleton.ca (Liang Qin) Date: Tue, 16 Mar 2004 11:07:06 -0500 Subject: [Xorp-hackers] is XORP server down? In-Reply-To: <200403152023.i2FKNgRj016391@possum.icir.org> References: <200403152023.i2FKNgRj016391@possum.icir.org> Message-ID: <4057262A.203@sce.carleton.ca> Cannot download from cvs server today. L. Qin From M.Handley@cs.ucl.ac.uk Tue Mar 16 16:13:18 2004 From: M.Handley@cs.ucl.ac.uk (Mark Handley) Date: Tue, 16 Mar 2004 16:13:18 +0000 Subject: [Xorp-hackers] is XORP server down? In-Reply-To: Your message of "Tue, 16 Mar 2004 11:07:06 EST." <4057262A.203@sce.carleton.ca> Message-ID: <55807.1079453598@aardvark.cs.ucl.ac.uk> >Cannot download from cvs server today. All of ICIR (where xorp.org is located) has been down for the last six hours or so, presumably due to a power outage. Berkeley seems to be suffering a lot recently. Everything seems to be back up OK now. I really need to look seriously at mirroring xorp.org here in London. Cheers, Mark From hodson@icir.org Tue Mar 16 16:15:03 2004 From: hodson@icir.org (Orion Hodson) Date: Tue, 16 Mar 2004 08:15:03 -0800 Subject: [Xorp-hackers] is XORP server down? In-Reply-To: Your message of "Tue, 16 Mar 2004 11:07:06 EST." <4057262A.203@sce.carleton.ca> Message-ID: <200403161615.i2GGF3E9030522@puma.icir.org> /-- Liang Qin wrote: | Cannot download from cvs server today. We've been hit by two unannounced power outages in the past 48 hours. We're back up again now. Kind Regards - Orion From terrys@zap.ncit.ca Tue Mar 16 16:58:57 2004 From: terrys@zap.ncit.ca (Terrance Sullivan) Date: Tue, 16 Mar 2004 11:58:57 -0500 Subject: [Xorp-hackers] is XORP server down? In-Reply-To: <55807.1079453598@aardvark.cs.ucl.ac.uk> References: <55807.1079453598@aardvark.cs.ucl.ac.uk> Message-ID: <200403161158.57514.terrys@zap.ncit.ca> Hi Mark, If you're thinking of mirroring Xorp to UCL, I'm wondering if it may also be useful here in Ottawa (i.e. Carleton and eastern NA in general) ? Certainly, if there's general consensus on this, I think NCIT would be willing to help make this happen. \Terry On March 16, 2004 11:13 am, Mark Handley wrote: > >Cannot download from cvs server today. > > All of ICIR (where xorp.org is located) has been down for the last six > hours or so, presumably due to a power outage. Berkeley seems to be > suffering a lot recently. Everything seems to be back up OK now. > > I really need to look seriously at mirroring xorp.org here in London. > > Cheers, > Mark From M.Handley@cs.ucl.ac.uk Tue Mar 16 17:12:42 2004 From: M.Handley@cs.ucl.ac.uk (Mark Handley) Date: Tue, 16 Mar 2004 17:12:42 +0000 Subject: [Xorp-hackers] is XORP server down? In-Reply-To: Your message of "Tue, 16 Mar 2004 11:58:57 EST." <200403161158.57514.terrys@zap.ncit.ca> Message-ID: <57590.1079457162@aardvark.cs.ucl.ac.uk> >If you're thinking of mirroring Xorp to UCL, I'm wondering if it may also be >useful here in Ottawa (i.e. Carleton and eastern NA in general) ? > >Certainly, if there's general consensus on this, I think NCIT would be willing >to help make this happen. Thanks for the offer! I think what we'll do is figure out how to do the UCL mirror first (I'm not clear at this point what the security story should be). Depending on how that works out, we may take you up on your offer. Cheers, Mark From lqin@sce.carleton.ca Tue Mar 16 18:39:46 2004 From: lqin@sce.carleton.ca (Liang Qin) Date: Tue, 16 Mar 2004 13:39:46 -0500 Subject: [Xorp-hackers] FEA is killed Message-ID: <405749F2.4020503@sce.carleton.ca> Hi XORP users, When I ran the XORP v0.5 and the latest code downloaded from CVS server (added my own module), I got this message: [ 2004/03/16 13:11:28 FATAL xorp_fea:4312 FEA +179 ifconfig_parse_nlm.cc nlm_newlink_to_fea_cfg ] Could not find interface name for interface index 4 childhandler: 17 pid=4312, status=6 process was killed with signal 6 Module abnormally killed: fea I have an ethernet and a 802.11b wireless card installed. What could be the problem? Thanks! Liang Qin From pavlin@icir.org Tue Mar 16 20:21:19 2004 From: pavlin@icir.org (Pavlin Radoslavov) Date: Tue, 16 Mar 2004 12:21:19 -0800 Subject: [Xorp-hackers] FEA is killed In-Reply-To: Message from Liang Qin of "Tue, 16 Mar 2004 13:39:46 EST." <405749F2.4020503@sce.carleton.ca> Message-ID: <200403162021.i2GKLJRj028278@possum.icir.org> > Hi XORP users, > > When I ran the XORP v0.5 and the latest code downloaded from CVS server > (added my > own module), I got this message: > > [ 2004/03/16 13:11:28 FATAL xorp_fea:4312 FEA +179 ifconfig_parse_nlm.cc nlm_newlink_to_fea_cfg > ] Could not find interface name for interface index 4 > childhandler: 17 > pid=4312, status=6 > process was killed with signal 6 > Module abnormally killed: fea > > I have an ethernet and a 802.11b wireless card installed. > > > What could be the problem? Liang, Can you be a bit more specific about the circumstances when this happen? E.g., does this happen when you plug-in the wireless card into the machine? Also, what Linux version are you running on that machine. Thanks, Pavlin From lqin@sce.carleton.ca Tue Mar 16 20:36:31 2004 From: lqin@sce.carleton.ca (L. Qin) Date: Tue, 16 Mar 2004 15:36:31 -0500 Subject: [Xorp-hackers] FEA is killed In-Reply-To: <200403162021.i2GKLJRj028278@possum.icir.org> References: <200403162021.i2GKLJRj028278@possum.icir.org> Message-ID: <4057654F.8090905@sce.carleton.ca> Hi Pavlin, The wireless card is plu-in before booting the pc. This error message appears after : ## commit_pass2_done ## commit seems successful MasterConfigTree::config_done: success commit was successful, finalizing... finalizing done MasterConfigTree::config_done returning The linux kernel is: 2.4.20-13.9.HOSTAP That's the same PC which could not pass the check process last Friday. Now I installed some rpm as requested, it seems work. But I got this kind of error message when checking: -PASS: test_stcp [ 2004/03/16 15:23:06 WARNING test_lemming XrlFinderTarget ] Handling method for finder/0.2/resolve_xrl failed: XrlCmdError 102 Command failed Xrl target is not enabled. ..... dozens of same messages [ 2004/03/16 15:23:23 WARNING test_lemming XrlFinderTarget ] Handling method for finder/0.2/resolve_xrl failed: XrlCmdError 102 Command failed Xrl target is not enabled. PASS: test_lemming Thanks! Liang >Can you be a bit more specific about the circumstances when this >happen? >E.g., does this happen when you plug-in the wireless card into the >machine? > >Also, what Linux version are you running on that machine. > >Thanks, >Pavlin > > > From pavlin@icir.org Tue Mar 16 21:01:44 2004 From: pavlin@icir.org (Pavlin Radoslavov) Date: Tue, 16 Mar 2004 13:01:44 -0800 Subject: [Xorp-hackers] FEA is killed In-Reply-To: Message from "L. Qin" of "Tue, 16 Mar 2004 15:36:31 EST." <4057654F.8090905@sce.carleton.ca> Message-ID: <200403162101.i2GL1iRj028600@possum.icir.org> > The wireless card is plu-in before booting the pc. > > This error message appears after : > > ## commit_pass2_done > ## commit seems successful > MasterConfigTree::config_done: success > commit was successful, finalizing... > finalizing done > MasterConfigTree::config_done returning If you don't mind, could you send me the "interfaces" section of your config.boot file and the output of "ifconfig -a". You can mask the particular values of the addresses, but please put a note whether the address values in config.boot are same as the corresponding addresses returned by ifconfig -a. Also, does the error happen if you try to reboot without the wireless card. > The linux kernel is: 2.4.20-13.9.HOSTAP > > That's the same PC which could not pass the check process last Friday. > Now I installed > some rpm as requested, it seems work. But I got this kind of error BTW, were those only the RPMs mentioned in BUILD_NOTES or did you have to install something else as well? > message when checking: > > -PASS: test_stcp > [ 2004/03/16 15:23:06 WARNING test_lemming XrlFinderTarget ] Handling > method for finder/0.2/resolve_xrl failed: XrlCmdError 102 Command failed > Xrl target is not enabled. > > ..... > dozens of same messages > [ 2004/03/16 15:23:23 WARNING test_lemming XrlFinderTarget ] Handling > method for finder/0.2/resolve_xrl failed: XrlCmdError 102 Command failed > Xrl target is not enabled. > PASS: test_lemming You can ignore those warnings because that program itself tests scenarios when the target is not available, etc. As long as the test program itself returns no error, then the test is OK. Pavlin From M.Handley@cs.ucl.ac.uk Tue Mar 16 21:00:29 2004 From: M.Handley@cs.ucl.ac.uk (Mark Handley) Date: Tue, 16 Mar 2004 21:00:29 +0000 Subject: [Xorp-hackers] FEA is killed In-Reply-To: Your message of "Tue, 16 Mar 2004 15:36:31 EST." <4057654F.8090905@sce.carleton.ca> Message-ID: <736.1079470829@cs.ucl.ac.uk> As the failure was in the FEA, the first starting point would seem to be to check that the FEA passes all the validation tests on this platform. I believe the FEA validation tests don't all run unless you run them as root (because they need to access priviledged kernel APIs). So a good starting point would seem to be to cd to the fea directory, and from there run "gmake check" as root. If all the tests pass, only then should we look Liang is doing differently. Cheers, Mark From pavlin@icir.org Wed Mar 17 07:07:59 2004 From: pavlin@icir.org (Pavlin Radoslavov) Date: Tue, 16 Mar 2004 23:07:59 -0800 Subject: [Xorp-hackers] FEA is killed In-Reply-To: Message from Mark Handley of "Tue, 16 Mar 2004 21:00:29 GMT." <736.1079470829@cs.ucl.ac.uk> Message-ID: <200403170707.i2H77xRj054456@possum.icir.org> > As the failure was in the FEA, the first starting point would seem to > be to check that the FEA passes all the validation tests on this > platform. I believe the FEA validation tests don't all run unless you > run them as root (because they need to access priviledged kernel > APIs). So a good starting point would seem to be to cd to the fea > directory, and from there run "gmake check" as root. > > If all the tests pass, only then should we look Liang is doing > differently. The FEA root-specific tests use a different Makefile, hence they can be run with "gmake -f MakefileRootCheck check" (as a root). Unfortunately, due to the nature of the tests, the tests require host-specific setup/entry. E.g., look into the beginning of fea/test_config_interface.sh which contains the host-specific setup (interface name to mess-up with, its expected index, Ethernet address, etc). I believe the problem is that the particular code in the FEA expects the kernel to provide some specific information about an existing network interface (the interface name), but that information is missing. Given that this particular code works on our Linux machine, my guess is that the missing information probably comes from the removable wireless Ethernet card (e.g., if the card is removable then the kernel doesn't fill-in the information?). Unfortunately, I don't have a laptop with Linux installed on it to verify that theory, but I will dig around. Once the reason is found the fix will be easy. Regards, Pavlin