If this bug also exists in Squeak (which I expect since it seems a VM problem), I suggest to report it there too (and create a Mantis entry). Adrian On Jul 21, 2009, at 13:49 , Stéphane Ducasse wrote:

Indeed. Mike is moving just right now but it is important that this bug gets addressed.

Stef

On Jul 21, 2009, at 10:41 AM, Damien Cassou wrote:

...

http://code.google.com/p/pharo/issues/detail?id=961

-- Damien Cassou http://damiencassou.seasidehosting.st

"Lambdas are relegated to relative obscurity until Java makes them popular by not having them." James Iry

_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

Again, I *think* the problem is too much code removed, and failure to protect with #useOldNetwork. It blows up on IPv4; it happens both on Windows and Linux. If anyone has counter examples or can otherwise show me wrong, please set me straight. Bill -----Original Message----- From: pharo-project-bounces@lists.gforge.inria.fr [mailto:pharo-project-bounces@lists.gforge.inria.fr] On Behalf Of Damien Cassou Sent: Tuesday, July 21, 2009 11:29 AM To: Pharo-project@lists.gforge.inria.fr Subject: Re: [Pharo-project] Show stopper network bug on Windows On Tue, Jul 21, 2009 at 12:56 PM, Adrian Lienhard<adi@netstyle.ch> wrote:

If this bug also exists in Squeak (which I expect since it seems a VM problem), I suggest to report it there too (and create a Mantis entry).

No, it's only Pharo. -- Damien Cassou http://damiencassou.seasidehosting.st "Lambdas are relegated to relative obscurity until Java makes them popular by not having them." James Iry _______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

Yes we do not have a windows at hand. can you let us know from which update the problem is related.

In this case the first step woud be to identify which update introduced the problem.

Suppose that a core #10200 image still works, you could do:

10200 to: 10384 do: [ :i | Utilities updateFromServerThroughUpdateNumber: i. NetNameResolver localHostAddress ]

And then see after which update number the first exception is thrown. I did not test this, but it should work.

Adrian

I think I have it willing to keep running now. I looked for the #inform: in the process browser, and missed it the first time - I needed to ask for the full stack, and there it was, a good long way down the list :) It is now sending text to the transcript, so cross your fingers. Bill -----Original Message----- From: pharo-project-bounces@lists.gforge.inria.fr [mailto:pharo-project-bounces@lists.gforge.inria.fr] On Behalf Of Schwab,Wilhelm K Sent: Tuesday, July 21, 2009 2:22 PM To: Pharo-project@lists.gforge.inria.fr Subject: Re: [Pharo-project] Show stopper network bug on Windows It's trying to run on a win2k3 box. I say trying because it insists on telling me (modally) how many updates it processed - every time. Is there a way to gag that? Bill -----Original Message----- From: pharo-project-bounces@lists.gforge.inria.fr [mailto:pharo-project-bounces@lists.gforge.inria.fr] On Behalf Of Stéphane Ducasse Sent: Tuesday, July 21, 2009 1:36 PM To: Pharo-project@lists.gforge.inria.fr Subject: Re: [Pharo-project] Show stopper network bug on Windows Yes we do not have a windows at hand. can you let us know from which update the problem is related.

In this case the first step woud be to identify which update introduced the problem.

Suppose that a core #10200 image still works, you could do:

10200 to: 10384 do: [ :i | Utilities updateFromServerThroughUpdateNumber: i. NetNameResolver localHostAddress ]

And then see after which update number the first exception is thrown. I did not test this, but it should work.

Adrian

_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

This is not quite working, but I suspect this is a defect in primHostNameSize, and will fall by loading VM maker and looking at the socket plugin. I do not know enough to say whether the primitive needs to be fixed, or simply avoided consistent with #useOldNetwork. -----Original Message----- From: pharo-project-bounces@lists.gforge.inria.fr [mailto:pharo-project-bounces@lists.gforge.inria.fr] On Behalf Of Schwab,Wilhelm K Sent: Tuesday, July 21, 2009 2:44 PM To: Pharo-project@lists.gforge.inria.fr Subject: Re: [Pharo-project] Show stopper network bug on Windows I think I have it willing to keep running now. I looked for the #inform: in the process browser, and missed it the first time - I needed to ask for the full stack, and there it was, a good long way down the list :) It is now sending text to the transcript, so cross your fingers. Bill -----Original Message----- From: pharo-project-bounces@lists.gforge.inria.fr [mailto:pharo-project-bounces@lists.gforge.inria.fr] On Behalf Of Schwab,Wilhelm K Sent: Tuesday, July 21, 2009 2:22 PM To: Pharo-project@lists.gforge.inria.fr Subject: Re: [Pharo-project] Show stopper network bug on Windows It's trying to run on a win2k3 box. I say trying because it insists on telling me (modally) how many updates it processed - every time. Is there a way to gag that? Bill -----Original Message----- From: pharo-project-bounces@lists.gforge.inria.fr [mailto:pharo-project-bounces@lists.gforge.inria.fr] On Behalf Of Stéphane Ducasse Sent: Tuesday, July 21, 2009 1:36 PM To: Pharo-project@lists.gforge.inria.fr Subject: Re: [Pharo-project] Show stopper network bug on Windows Yes we do not have a windows at hand. can you let us know from which update the problem is related.

In this case the first step woud be to identify which update introduced the problem.

Suppose that a core #10200 image still works, you could do:

10200 to: 10384 do: [ :i | Utilities updateFromServerThroughUpdateNumber: i. NetNameResolver localHostAddress ]

And then see after which update number the first exception is thrown. I did not test this, but it should work.

Adrian

_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

Sig, Queue veryLoudHeadSlap.wav - however, that is not a fix for it on Linux. Also, on Windows, I am not getting the fully qualified name, which is another problem. It also helps make the case for formalized session management, as the image should know when it has had its first cup of coffee (the vm can inform it). Bill -----Original Message----- From: pharo-project-bounces@lists.gforge.inria.fr [mailto:pharo-project-bounces@lists.gforge.inria.fr] On Behalf Of Igor Stasenko Sent: Tuesday, July 21, 2009 6:53 PM To: Pharo-project@lists.gforge.inria.fr Subject: Re: [Pharo-project] Show stopper network bug on Windows Guys, see my comment on that issue.: #localHostName should not be used before #initializeNetwork. Hope it helps. _______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

Sig, Respectfully, I do not agree that the current situation is elegant. There are other ways to cope with a disabled plugin (see #useOldNetwork), and littering code with #initializeNetwork is error prone and wasteful in its own way, no matter how lean the method might be. This looks to me like an anachronism. There was no better mechanism, so a workaround had to suffice. Now with weak collections and events, it should be possible to do better. If those things do not work well, then they need to be challenged and fixed as appropriate. The lack of fully qualified names could easily be a windows stupidity. My Dolphin code for lookups has long been a little complicated: look up this, clear that, do it again, and it always seemed that I was fighting lack of imagination from the general direction of Redmond WA - who would have thought that possible? :) My favorite was win2k from (IIRC) pre-sp3: the only machine one could not find by name was the local host! Bill -----Original Message----- From: pharo-project-bounces@lists.gforge.inria.fr [mailto:pharo-project-bounces@lists.gforge.inria.fr] On Behalf Of Igor Stasenko Sent: Tuesday, July 21, 2009 11:02 PM To: Pharo-project@lists.gforge.inria.fr Subject: Re: [Pharo-project] Show stopper network bug on Windows 2009/7/22 Schwab,Wilhelm K <bschwab@anest.ufl.edu>:

Sig,

Queue veryLoudHeadSlap.wav - however, that is not a fix for it on Linux.  Also, on Windows, I am not getting the fully qualified name, which is another problem.  It also helps make the case for formalized session management, as the image should know when it has had its first cup of coffee (the vm can inform it).

Usually, all code , working with sockets , sending #initializeNetwork, to be sure it is initialized: Socket>>newTCP: family "Create a socket and initialise it for TCP" self initializeNetwork. ^[ super new initialize: TCPSocketType family: family ] repeatWithGCIf: [ :socket | socket isValid not ] --------- Socket>>initializeNetwork "Initialize the network drivers and the NetNameResolver. Do nothing if the network is already initialized." "Note: The network must be re-initialized every time Squeak starts up, so applications that persist across snapshots should be prepared to re-initialize the network as needed. Such applications should call 'Socket initializeNetwork' before every network transaction. " NetNameResolver initializeNetwork ---------- another way could be just add the #startUp method and initialize network there. But i think that current solution is more elegant, because if your image never using the network, it will never initialize the plugin and in this way, preserve some resources. And , of course , there could be the cases, where VMs built without network support, or running with disabled networking security setting (for sandboxing). In these cases this primitive (as well as rest of socket prims) will fail. As for retreiving a fully qualified name - can't give you answer right now. I think it can be answered by Andreas, who added the ipv6 support recently to Windoze VM platform code.

Bill

-----Original Message----- From: pharo-project-bounces@lists.gforge.inria.fr [mailto:pharo-project-bounces@lists.gforge.inria.fr] On Behalf Of Igor Stasenko Sent: Tuesday, July 21, 2009 6:53 PM To: Pharo-project@lists.gforge.inria.fr Subject: Re: [Pharo-project] Show stopper network bug on Windows

Guys, see my comment on that issue.: #localHostName should not be used before #initializeNetwork.

Hope it helps.

_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

-- Best regards, Igor Stasenko AKA sig. _______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

This is a squeakdebug.log lines: (again, it fails for me only if i never issue the #initializeNetwork after image startup before using #localHostName) ------- Error: a primitive has failed 22 July 2009 10:41:57 pm VM: Win32 - IX86 - NT - Squeak3.10.2 of '5 June 2008' [latest update: #7179] Image: Pharo0.1 [Latest update: #10371] SecurityManager state: Restricted: false FileAccess: true SocketAccess: true Working Dir **** Trusted Dir **** Untrusted Dir **** NetNameResolver class(Object)>>error: Receiver: NetNameResolver Arguments and temporary variables: aString: 'a primitive has failed' ... <snip> ... NetNameResolver class>>primHostNameSize -------------- 2009/7/22 Igor Stasenko <siguctua@gmail.com>:

Okay, then i think it is better to move this thread on vm-dev list (to notify VM devs about this bug), because it seems a VM bug, not image-side bug.

I just tested it on my WinXP + VM (dated 27 march 2009 - not sure what version is) and it indeed failing if i do NetNameResolver localHostName and not do: NetNameResolver initializeNetwork before.

But once i did initialization, all subsequent requests is ok.

2009/7/22 Schwab,Wilhelm K <bschwab@anest.ufl.edu>:

...

Sig,

Respectfully, I do not agree that the current situation is elegant.  There are other ways to cope with a disabled plugin (see #useOldNetwork), and littering code with #initializeNetwork is error prone and wasteful in its own way, no matter how lean the method might be.

This looks to me like an anachronism.  There was no better mechanism, so a workaround had to suffice.  Now with weak collections and events, it should be possible to do better.  If those things do not work well, then they need to be challenged and fixed as appropriate.

The lack of fully qualified names could easily be a windows stupidity.  My Dolphin code for lookups has long been a little complicated: look up this, clear that, do it again, and it always seemed that I was fighting lack of imagination from the general direction of Redmond WA - who would have thought that possible? :)  My favorite was win2k from (IIRC) pre-sp3: the only machine one could not find by name was the local host!

Bill

-----Original Message----- From: pharo-project-bounces@lists.gforge.inria.fr [mailto:pharo-project-bounces@lists.gforge.inria.fr] On Behalf Of Igor Stasenko Sent: Tuesday, July 21, 2009 11:02 PM To: Pharo-project@lists.gforge.inria.fr Subject: Re: [Pharo-project] Show stopper network bug on Windows

2009/7/22 Schwab,Wilhelm K <bschwab@anest.ufl.edu>:

...

Sig,

Queue veryLoudHeadSlap.wav - however, that is not a fix for it on Linux.  Also, on Windows, I am not getting the fully qualified name, which is another problem.  It also helps make the case for formalized session management, as the image should know when it has had its first cup of coffee (the vm can inform it).

Usually, all code , working with sockets , sending #initializeNetwork, to be sure it is initialized:

Socket>>newTCP: family        "Create a socket and initialise it for TCP"        self initializeNetwork.        ^[ super new initialize: TCPSocketType family: family ]                repeatWithGCIf: [ :socket | socket isValid not ]

--------- Socket>>initializeNetwork        "Initialize the network drivers and the NetNameResolver. Do nothing if the network is already initialized."        "Note: The network must be re-initialized every time Squeak starts up, so applications that persist across snapshots should be prepared to re-initialize the network as needed. Such applications should call 'Socket initializeNetwork' before every network transaction. "

       NetNameResolver initializeNetwork ---------- another way could be just add the #startUp method and initialize network there. But i think that current solution is more elegant, because if your image never using the network, it will never initialize the plugin and in this way, preserve some resources. And , of course , there could be the cases, where VMs built without network support, or running with disabled networking security setting (for sandboxing). In these cases this primitive (as well as rest of socket prims) will fail.

As for retreiving a fully qualified name - can't give you answer right now. I think it can be answered by Andreas, who added the ipv6 support recently to Windoze VM platform code.

...

Bill

-----Original Message----- From: pharo-project-bounces@lists.gforge.inria.fr [mailto:pharo-project-bounces@lists.gforge.inria.fr] On Behalf Of Igor Stasenko Sent: Tuesday, July 21, 2009 6:53 PM To: Pharo-project@lists.gforge.inria.fr Subject: Re: [Pharo-project] Show stopper network bug on Windows

Guys, see my comment on that issue.: #localHostName should not be used before #initializeNetwork.

Hope it helps.

_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

-- Best regards, Igor Stasenko AKA sig.

_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

-- Best regards, Igor Stasenko AKA sig.

-- Best regards, Igor Stasenko AKA sig.

2009/7/22 Schwab,Wilhelm K <bschwab@anest.ufl.edu>:

The lack of fully qualified names could easily be a windows stupidity.  My Dolphin code for lookups has long been a little complicated: look up this, clear that, do it again, and it always seemed that I was fighting lack of imagination from the general direction of Redmond WA - who would have thought that possible? :)  My favorite was win2k from (IIRC) pre-sp3: the only machine one could not find by name was the local host!

Concerning fully qualified host name. Yes. This is a well-known issue of windoze. For instance, my machine returns 'comp' on #localHostName request. Which is nothing more, than a computer name, i assigned in windoze settings. I didn't assigned my PC to be a part of any (sub)domain, so this is all it can report. The problems with dns/windoze host naming is comes back to Windows 3.11 era when there was no 'internet' but a local networks and windows used own identification mechanisms and protocols (NetBIOS, NetBEUI) to connect different hosts into a 'workgroup' :) However, next is getting interesting. I having a following network interfaces on my PC (i have to translate the text, because it prints things in russian) : (( The engineers at Redmond renamed the well-known unix ifconfig utility to ipconfig :) )) C:\>ipconfig /all Windows IP protocol settings Host name . . . . . . . . . : comp Primary DNS-suffix . . . . . . : Host type . . . . . . . . . . . . . : unknown IP-forwarding enabled . . . . : no WINS-proxy enabled . . . . . . . : no VMware Network Adapter VMnet8 - Ethernet adapter: DNS-suffix . . : IP-address . . . . . . . . . . . . : 192.168.6.1 Subnet mask . . . . . . . . . . : 255.255.255.0 Gateway . . . . . . . . . . : VMware Network Adapter VMnet1 - Ethernet adapter: DNS-suffix . . : IP-address . . . . . . . . . . . . : 192.168.89.1 Subnet mask . . . . . . . . . . : 255.255.255.0 Gateway . . . . . . . . . . : VirtualBox Host-Only Network - Ethernet adapter: DNS-suffix . . : IP-address . . . . . . . . . . . . : 192.168.56.1 Subnet mask . . . . . . . . . . : 255.255.255.0 Gateway . . . . . . . . . . : Network bridge - Ethernet adapter: DNS-suffix . . : IP-address . . . . . . . . . . . . : 192.168.0.3 Subnet mask . . . . . . . . . . : 255.255.255.0 Gateway . . . . . . . . . . : 192.168.0.1 Now, if i run: C:\>ping comp it starts pinging 192.168.56.1 address, which its thinks my extenal/local adapter address, but its obviously not i would expect that 'comp' should be resolved to 127.0.0.1 address (a loopback adapter) but windows thinks differently :) So, imagine i want to create a socket , listening for incoming connections at 'localHostName' address: NetNameResolver addressForName: NetNameResolver localHostName it gives me the same result as ping does: 192.168.56.1(192.168.56.1),0(0) surely, using this address i will never receive any incoming connections.. :) But what i didn't expected, it also opens a debug console and prints some kind of error about getnameinfo() function (again, translated, because my windoze is localized): getnameinfo: requested name is allowed and its found in database, but for this name the associated data is missing, which was allowed for it.. But its not producing a primitive failure. btw, same error shown in concole even if i do: NetNameResolver addressForName: 'localhost' i found that the error producing when i printing the instance of SocketAddress (#hostName method) weird... Next, nslookup utility of course, unable to resolve 'comp' name: C:\>nslookup Default Server: ns.ip.net.ua Address: 82.193.96.6

comp Server: ns.ip.net.ua Address: 82.193.96.6

*** ns.ip.net.ua can't find comp: Non-existent domain

-- Best regards, Igor Stasenko AKA sig.

Ironically, it did for me on windows, but not on Linux. -----Original Message----- From: pharo-project-bounces@lists.gforge.inria.fr [mailto:pharo-project-bounces@lists.gforge.inria.fr] On Behalf Of Damien Cassou Sent: Wednesday, July 22, 2009 3:36 AM To: Pharo-project@lists.gforge.inria.fr Subject: Re: [Pharo-project] Show stopper network bug on Windows On Wed, Jul 22, 2009 at 12:53 AM, Igor Stasenko<siguctua@gmail.com> wrote:

Guys, see my comment on that issue.: #localHostName should not be used before #initializeNetwork.

#initializeNetwork doesn't prevent the bug from happening. -- Damien Cassou http://damiencassou.seasidehosting.st "Lambdas are relegated to relative obscurity until Java makes them popular by not having them." James Iry _______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project