Nor here.

No related posts.

Doesn’t look too bad from the front, does it!

The stretch Ute.

No need for a door anymore.

Always turns to the right a bit, though.

Nice and cool on a hot day.

The dash was a little bit warped, though. It could have come out better.

No related posts.

Enter Solaris 8 brandz zones. Being able to run legacy applications in a Solaris 8 brandz generally gets you around most of the issues in the remaining apps that do stupid things. However, there’s a couple of quirks with a zone that can introduce odd behaviour in an application.

Progress for example……. Read On.

The issue

An old legacy Progress database was sitting on an old legacy server which was so EOL that it would make even Bill cringe… or laugh… Running Solaris 6 was not an option, but running anything else, (even Solaris 9), would be adventurous to say the least. Can I say legacy again?

The solution

So first try a brandz Solaris 8. Configured, installed, and booted a brand spanking new zone. Copy the data over. Attempt a start of the Progress database.

10:03:21 BROKER  0: Multi-user session begin. (333)
10:03:24 BROKER  0: Begin Physical Redo Phase at 13312 . (5326)
10:03:30 BROKER  0: Physical Redo Phase Completed at blk 13743 off 2457 upd 2636. (7161)
10:03:31 BROKER  0: Started for mfglive using TCP, pid 20533. (5644)
10:04:02 BROKER  0: SYSTEM ERROR: Unable to kill parent process, errno= 4. (1680)
10:04:02 BROKER  0: SYSTEM ERROR: The broker is exiting unexpectedly, beginning Abnormal Shutdown. (5292)
10:04:02 BROKER  0: drexit: Initiating Abnormal Shutdown
10:04:02 BROKER  0: ** Save file named core for analysis by Progress Software Corporation. (439)
10:04:02 BROKER  0: Begin ABNORMAL shutdown code 2 (2249)
10:04:04 BROKER   : Removed shared memory with segment_id: 41
10:04:04 BROKER   : Multi-user session end. (334)

Nada. Mmmmm. Seems to start up OK, but then crashes in a screaming heap a short while later.

Checking shared libraries seemed to be OK. Nothing missing or odd.

root % ldd -r /mfgpro/dlc91c/bin/_mprosrv
        /usr/lib/secure/s8_preload.so.1
        libsocket.so.1 =>        /usr/lib/libsocket.so.1
        libnsl.so.1 =>   /usr/lib/libnsl.so.1
        libintl.so.1 =>  /usr/lib/libintl.so.1
        libdl.so.1 =>    /usr/lib/libdl.so.1
        libm.so.1 =>     /usr/lib/libm.so.1
        libthread.so.1 =>        /usr/lib/libthread.so.1
        libc.so.1 =>     /usr/lib/libc.so.1
        libmp.so.2 =>    /usr/lib/libmp.so.2
        /usr/platform/SUNW,Sun-Fire-V490/lib/libc_psr.so.1

So, I thought I’d truss, (the old friend), a few processes to see what was going on. Chucking the below in various scripts, and comparing the working server to the non-working server. (This produces VERY NOISY output, but it’s great for diagnosing what’s going on.)

truss -feal -vall -xall -rall -wall -sall -mall -o /tmp/logfile exec_process

If you look at the appendices to this post, I’ve added the relevant, sections of the truss. You may notice an interesting thing. The working server roughly follows this process, (in the snippet):
- Bind to 0.0.0.0 port 3500.
- fstat64() filehandles.
- Yadda, yadda, write output to STDOUT.
- getpid() – find parent process, (24012).
- kill -TERM 24012
- wait()
- Parent – ALRM signal
- Parent – die.
- Child – Context switch.
- getpid() – find parent process, (1). OK!

Now this is a little odd. It would seem that some developer thought that it’d be a damn good idea to re-invent the old “I’m going to fork a child process in the background” API calls, and chuck in their own.

They are checking the parent process, killing it, and rechecking to see if it’s now 1, (which is sysvinit), then it progresses on.

Now, the non-working server.
- Bind to 0.0.0.0 port 3500.
- fstat64() filehandles.
- Yadda, yadda, write output to STDOUT.
- getpid() – find parent process, (21267).
- kill -TERM 21267
- wait()
- Parent – ALRM signal <-IMPORTANT
- Parent - no die.
- getpid() - find parent process, (21267).
- kill -TERM 21267
- wait()
- Parent - ALRM signal <-IMPORTANT
- Parent - no die.
Ad naseum.....

A quick process check:

     UID   PID  PPID  C    STIME TTY      TIME CMD
    root  2260  1417  0   May 27 ?        0:09 /etc/init
    root  1417  1417  0   May 27 ?        0:00 zsched
    root  2960  1417  0   May 27 ?        0:00 /usr/lib/picl/picld

Hey! Of course you're not gonna be able to kill zsched! You see in a Solaris zone the 'root' process is called 'zsched', and it will never ever be 1 at all, and neither will init for that matter. Not only that you're not ever going to be able to kill the zsched process from within the zone as it's protected within the global zone memory space.

Well, it's too late to kick the developer up the bum. So time to hack.

The hack

Preloading shared libs has been around since I was a Mars Bar in my Dad's top pocket, (well almost). So it should be a simple case of replacing any getpid() API call with my own. Simple enough, done it before.

See the appendix for the getpidhack.c file that I used. This is a shared lib that replaces both getppid() and getpid(), (for good measure, but not really needed in my case), API calls. It will return the PID 1, when _getppid() returns the same PID as the environment variable PIDHACK.
Note that I call _getppid() which allows me to fetch the REAL PID so I can return it when they don't match.

So, to use it. First build it:

gcc -fPIC -c getpidhack.c
gld getpidhack.o -L/usr/lib -lc -G -assert pure-text -o /usr/local/lib/libgetpidhack.so

Set an environment variable called PIDHACK to the currently running zsched process:

HACKPID=`ps -e | awk '/zsched/{print$1}'`
export HACKPID

Then set your LD_PRELOAD variable to point to the shared lib:

LD_PRELOAD="/usr/local/lib/libgetpidhack.so"
export LD_PRELOAD

After this the Progress database just started. Woot! The DB can now run on hardware that it would never in it's life be able to run on under a brandz zone. Another EOL server to chuck out the computer room window to see what happens.

BTW, if you are having issues with this, you can run a DTrace to see if you're calling the real getpid(), or the getpidhack version. Create the following script, and run it with one argument, (the PID of your zsched). Every time you hit the getppid() API call it'll register - this means that you are not replacing the API call and your shared lib preload isn't working.

#!/usr/sbin/dtrace -s

syscall::getpid:return
/ppid == $1/
{
        printf("%d %d", ppid, pid);
}

Appendix 1 - getpidhack.c

#include <sys/types.h>
#include <unistd.h>
#include <stdio.h>

pid_t getppid(void)
{
        pid_t hackppid;
        pid_t realppid;

        // hackpid - the PID of the zsched process.
        hackppid = atol(getenv("HACKPID"));

        // realppid - the real PID of the parent process.
        realppid = _getppid();

        // If the realppid equals hackppid, then we want to reset it to 1.
        if (realppid == hackppid)
        {
                return((pid_t) 1);
        }

        // Else return whatever PID we were given.
        return(realppid);
}

pid_t getpid(void)
{
        pid_t hackpid;
        pid_t realpid;

        // hackpid - the PID of the zsched process.
        hackpid = atol(getenv("HACKPID"));

        // realpid - the real PID of the parent process.
        realpid = _getpid();

        // If the realpid equals hackpid, then we want to reset it to 1.
        if (realpid == hackpid)
        {
                return((pid_t) 1);
        }

        // Else return whatever PID we were given.
        return(realpid);
}

Appendix 2 - The working server

24013/1:        setsockopt(76, 65535, 4, 0xEFFFF7FC, 4)         = 0
24013/1:        setsockopt(76, 65535, 8, 0xEFFFF7FC, 4)         = 0
24013/1:        bind(76, 0x0015DCA8, 16)                        = 0
24013/1:                name = 0.0.0.0/3500
24013/1:        listen(76, 10)                                  = 0
24013/1:        getpid()                                        = 24013 [24012]
24013/1:        lseek(3, 457164, 0)                             = 457164
24013/1:        read(3, 0xEFFFF66C, 81)                         = 81
24013/1:           % L S t a r t e d   f o r   % s   u s i n g   % s ,   p i d  %
24013/1:           l .   ( 5 6 4 4 )\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0 \0
24013/1:          \0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0
24013/1:        time()                                          = 1252542477
24013/1:        write(9, 0xEFFFF154, 69)                        = 69
24013/1:           1 0 : 2 7 : 5 7   B R O K E R     0 :   S t a r t e d   f o r
24013/1:           m f g l i v e   u s i n g   T C P ,   p i d   2 4 0 1 3 .   ( 5
24013/1:           6 4 4 )\n
24013/1:        fstat64(0, 0xEFFFF058)                          = 0
24013/1:            d=0x01540000 i=17139 m=0020620 l=1  u=0     g=7     rdev=0x0 0600005
24013/1:                at = Sep 10 10:27:46 EST 2009  [ 1252542466 ]
24013/1:                mt = Sep 10 10:27:56 EST 2009  [ 1252542476 ]
24013/1:                ct = Sep 10 09:39:31 EST 2009  [ 1252539571 ]
24013/1:            bsz=8192  blks=0     fs=ufs
24013/1:        fstat64(1, 0xEFFFF058)                          = 0
24013/1:            d=0x01540000 i=17139 m=0020620 l=1  u=0     g=7     rdev=0x0 0600005
24013/1:                at = Sep 10 10:27:46 EST 2009  [ 1252542466 ]
24013/1:                mt = Sep 10 10:27:56 EST 2009  [ 1252542476 ]
24013/1:                ct = Sep 10 09:39:31 EST 2009  [ 1252539571 ]
24013/1:            bsz=8192  blks=0     fs=ufs
24013/1:        fstat64(2, 0xEFFFF058)                          = 0
24013/1:            d=0x01540000 i=17139 m=0020620 l=1  u=0     g=7     rdev=0x0 0600005
24013/1:                at = Sep 10 10:27:46 EST 2009  [ 1252542466 ]
24013/1:                mt = Sep 10 10:27:56 EST 2009  [ 1252542476 ]
24013/1:                ct = Sep 10 09:39:31 EST 2009  [ 1252539571 ]
24013/1:            bsz=8192  blks=0     fs=ufs
24013/1:        write(1, 0xEFFFF154, 69)                        = 69
24013/1:           1 0 : 2 7 : 5 7   B R O K E R     0 :   S t a r t e d   f o r
24013/1:           m f g l i v e   u s i n g   T C P ,   p i d   2 4 0 1 3 .   ( 5
24013/1:           6 4 4 )\n
24013/1:        pwrite64(7, 0xEFFFF908, 4, 0)                   = 4
24013/1:             0xEFFFF908: "\0\0\002"
24013/1:        close(7)                                        = 0
24013/1:        getpid()                                        = 24013 [24012]
24013/1:        getpid()                                        = 24013 [24012]
24013/1:        kill(24012, 0x0000000F)                         = 0
24012/2:        signotifywait()                                 = 15
24013/1:        lwp_alarm(1)                                    = 0
24012/2:            Incurred fault #11, FLTPAGE  %pc = 0xEF5CDC80  addr = 0xEF67 0F48
24012/2:            Incurred fault #11, FLTPAGE  %pc = 0xEF64A3C8  addr = 0xEF66 FA68
24012/2:        lwp_sigredirect(1, 0x0000000F)                  = 0
24012/1:            Received signal #15, SIGTERM, in wait() [caught]
24012/1:              siginfo: SIGTERM pid=24013 uid=0
24012/1:        wait()                                          Err#4 EINTR
24012/1:        sigprocmask(3, 0xEF667DF8, 0x00000000)          = 0
24012/1:                 set = 0 0 0 0
24012/1:            Incurred fault #11, FLTPAGE  %pc = 0x0004593C  addr = 0x0013 10F4
24012/1:            Incurred fault #11, FLTPAGE  %pc = 0x000459A8  addr = 0x0015 3AEC
24012/1:        _exit(0)
24013/1:            Received signal #14, SIGALRM, in pause() [caught]
24013/1:        pause()                                         Err#4 EINTR
24013/1:            Incurred fault #11, FLTPAGE  %pc = 0xEF5CDCE8  addr = 0xEF5C DCE8
24013/1:        sigprocmask(3, 0xEF667DF8, 0x00000000)          = 0
24013/1:                 set = 0 0 0 0
24013/1:        context(1, 0xEFFFF240)
24013/1:        getpid()                                        = 24013 [1]
24013/1:        close(0)                                        = 0
24013/1:        open(0xEFFFF920, 0)                             = 0
24013/1:             0xEFFFF920: "/qaddb/eblive/mfglive.lg"
24013/1:        close(1)                                        = 0
24013/1:        open(0xEFFFF920, 0)                             = 1
24013/1:             0xEFFFF920: "/qaddb/eblive/mfglive.lg"
24013/1:        close(2)                                        = 0
24013/1:        open(0xEFFFF920, 0)                             = 2
24013/1:             0xEFFFF920: "/qaddb/eblive/mfglive.lg"
24013/1:        lseek(3, 342954, 0)                             = 342954
24013/1:        read(3, 0xEFFFF570, 81)                         = 81
24013/1:           % L P R O G R E S S   V e r s i o n   % s   o n   % s .   ( 4 2
24013/1:           3 4 )\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0 \0
24013/1:          \0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0
24013/1:        time()                                          = 1252542478
24013/1:        write(9, 0xEFFFF05C, 61)                        = 61
24013/1:           1 0 : 2 7 : 5 8   B R O K E R     0 :   P R O G R E S S   V e r
24013/1:           s i o n   9 . 1 C   o n   S O L A R I S .   ( 4 2 3 4 )\n
24013/1:        fstat64(0, 0xEFFFEF60)                          = 0
24013/1:            d=0x01540002 i=68332 m=0100644 l=1  u=0     g=1     sz=97192 88
24013/1:                at = Sep 10 03:01:28 EST 2009  [ 1252515688 ]
24013/1:                mt = Sep 10 10:27:58 EST 2009  [ 1252542478 ]
24013/1:                ct = Sep 10 10:27:58 EST 2009  [ 1252542478 ]
24013/1:            bsz=8192  blks=19008 fs=ufs
24013/1:        fstat64(1, 0xEFFFEF60)                          = 0
24013/1:            d=0x01540002 i=68332 m=0100644 l=1  u=0     g=1     sz=97192 88
24013/1:                at = Sep 10 03:01:28 EST 2009  [ 1252515688 ]
24013/1:                mt = Sep 10 10:27:58 EST 2009  [ 1252542478 ]
24013/1:                ct = Sep 10 10:27:58 EST 2009  [ 1252542478 ]
24013/1:            bsz=8192  blks=19008 fs=ufs
24013/1:        fstat64(2, 0xEFFFEF60)                          = 0
24013/1:            d=0x01540002 i=68332 m=0100644 l=1  u=0     g=1     sz=97192 88
24013/1:                at = Sep 10 03:01:28 EST 2009  [ 1252515688 ]
24013/1:                mt = Sep 10 10:27:58 EST 2009  [ 1252542478 ]
24013/1:                ct = Sep 10 10:27:58 EST 2009  [ 1252542478 ]
24013/1:            bsz=8192  blks=19008 fs=ufs
24013/1:        write(1, 0xEFFFF05C, 61)                        Err#9 EBADF
24013/1:           1 0 : 2 7 : 5 8   B R O K E R     0 :   P R O G R E S S   V e r
24013/1:           s i o n   9 . 1 C   o n   S O L A R I S .   ( 4 2 3 4 )\n
24013/1:        lseek(3, 346761, 0)                             = 346761
24013/1:        read(3, 0xEFFFF570, 81)                         = 81
24013/1:           % L S e r v e r   s t a r t e d   b y   % s   o n   % s .   ( 4
24013/1:           2 8 1 )\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0 \0
24013/1:          \0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0
24013/1:        time()                                          = 1252542478
24013/1:        write(9, 0xEFFFF05C, 65)                        = 65
24013/1:           1 0 : 2 7 : 5 8   B R O K E R     0 :   S e r v e r   s t a r t
24013/1:           e d   b y   r o o t   o n   / d e v / p t s / 5 .   ( 4 2 8 1)
24013/1:          \n

Appendix 3 - The non-working server

21268/1:        setsockopt(76, 65535, 4, 0xFFBFF9AC, 4, 1)      = 0
21268/1:        setsockopt(76, 65535, 8, 0xFFBFF9AC, 4, 1)      = 0
21268/1:        bind(76, 0x0015DCA8, 16, 3)                     = 0
21268/1:                AF_INET  name = 0.0.0.0  port = 3500
21268/1:        listen(76, 10, 1)                               = 0
21268/1:        getpid()                                        = 21268 [21267]
21268/1:        lseek(3, 457164, 0)                             = 457164
21268/1:        read(3, 0xFFBFF81C, 81)                         = 81
21268/1:           % L S t a r t e d   f o r   % s   u s i n g   % s ,   p i d   %
21268/1:           l .   ( 5 6 4 4 )\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0
21268/1:          \0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0
21268/1:        time()                                          = 1252542489
21268/1:        write(9, 0xFFBFF304, 69)                        = 69
21268/1:           1 0 : 2 8 : 0 9   B R O K E R     0 :   S t a r t e d   f o r
21268/1:           m f g l i v e   u s i n g   T C P ,   p i d   2 1 2 6 8 .   ( 5
21268/1:           6 4 4 )\n
21268/1:        fstat64(0, 0xFFBFF208)                          = 0
21268/1:            d=0x05980009 i=4426329 m=0020620 l=1  u=0     g=7     rdev=0x00600011
21268/1:                at = Sep 10 10:27:57 EST 2009  [ 1252542477 ]
21268/1:                mt = Sep 10 10:28:09 EST 2009  [ 1252542489 ]
21268/1:                ct = Sep 10 09:44:02 EST 2009  [ 1252539842 ]
21268/1:            bsz=8192  blks=0     fs=lofs
21268/1:        fstat64(1, 0xFFBFF208)                          = 0
21268/1:            d=0x05980009 i=4426329 m=0020620 l=1  u=0     g=7     rdev=0x00600011
21268/1:                at = Sep 10 10:27:57 EST 2009  [ 1252542477 ]
21268/1:                mt = Sep 10 10:28:09 EST 2009  [ 1252542489 ]
21268/1:                ct = Sep 10 09:44:02 EST 2009  [ 1252539842 ]
21268/1:            bsz=8192  blks=0     fs=lofs
21268/1:        fstat64(2, 0xFFBFF208)                          = 0
21268/1:            d=0x05980009 i=4426329 m=0020620 l=1  u=0     g=7     rdev=0x00600011
21268/1:                at = Sep 10 10:27:57 EST 2009  [ 1252542477 ]
21268/1:                mt = Sep 10 10:28:09 EST 2009  [ 1252542489 ]
21268/1:                ct = Sep 10 09:44:02 EST 2009  [ 1252539842 ]
21268/1:            bsz=8192  blks=0     fs=lofs
21268/1:        write(1, 0xFFBFF304, 69)                        = 69
21268/1:           1 0 : 2 8 : 0 9   B R O K E R     0 :   S t a r t e d   f o r
21268/1:           m f g l i v e   u s i n g   T C P ,   p i d   2 1 2 6 8 .   ( 5
21268/1:           6 4 4 )\n
21268/1:        pwrite64(7, 0xFFBFFAB8, 4, 0)                   = 4
21268/1:             0xFFBFFAB8: "\0\0\002"
21268/1:        close(7)                                        = 0
21268/1:        getpid()                                        = 21268 [21267]
21268/1:        kill(21267, 0x0000000F)                         = 0
21267/1:            Received signal #15, SIGTERM, in wait() [caught]
21267/1:              siginfo: SIG#0
21268/1:            Incurred fault #11, FLTPAGE  %pc = 0x000C3048  addr = 0x58900137234
21267/1:        wait()                                          Err#4 EINTR
21268/1:        alarm(1)                                        = 0
21267/1:        lwp_sigtimedwait(0xFFBFF570, 0xFFBFF498, 0x00000010) = 0
21267/1:                sigmask = 0 0 0 0
21267/1:              siginfo: SIG#0
21267/1:        lwp_sigtimedwait(0xFFBFF488, 0xFFBFF570, 0x00000010) = 0
21267/1:                sigmask = 0 0 0 0
21267/1:              siginfo: SIG#0
21267/1:        lwp_sigtimedwait(0xFFBFF484, 0xFFBFF2B8, 0x00000010) = 0
21267/1:                sigmask = 0x00004000 0 0 0
21267/1:              siginfo: SIG#16384
21267/1:        lwp_sigtimedwait(0xFFBFF2A8, 0xFFBFF330, 0x00000010) = 0
21267/1:                sigmask = 0x00004000 0 0 0
21267/1:              siginfo: SIG#16384
21267/1:        sigprocmask(3, 0xFFBFF330, 0x00000000)          = 0
21267/1:                 set = 0x00004000 0 0 0
21267/1:            Incurred fault #11, FLTPAGE  %pc = 0x0004593C  addr = 0x589001310F4
21267/1:            Incurred fault #11, FLTPAGE  %pc = 0x000459A8  addr = 0x58900153AEC
21267/1:        _exit(0)
21268/1:            Received signal #14, SIGALRM, in pause() [caught]
21268/1:              siginfo: SIG#0
21268/1:        pause()                                         Err#4 EINTR
21268/1:        lwp_sigtimedwait(0xFFBFF698, 0xFFBFF5C0, 0x00000010) = 0
21268/1:                sigmask = 0 0 0 0
21268/1:              siginfo: SIG#0
21268/1:        lwp_sigtimedwait(0xFFBFF5B0, 0xFFBFF698, 0x00000010) = 0
21268/1:                sigmask = 0 0 0 0
21268/1:              siginfo: SIG#0
21268/1:        lwp_sigtimedwait(0xFFBFF5AC, 0xFFBFF3E0, 0x00000010) = 0
21268/1:                sigmask = 0x00002000 0 0 0
21268/1:              siginfo: SIG#8192
21268/1:        lwp_sigtimedwait(0xFFBFF3D0, 0xFFBFF458, 0x00000010) = 0
21268/1:                sigmask = 0x00002000 0 0 0
21268/1:              siginfo: SIG#8192
21268/1:        sigprocmask(3, 0xFFBFF458, 0x00000000)          = 0
21268/1:                 set = 0x00002000 0 0 0
21268/1:        lwp_sigtimedwait(0xFF1D7C18, 0xFFBFF1B0, 0x00000010) = 0
21268/1:                sigmask = 0xFFBFFEFF 0x00001FFF 0 0
21268/1:              siginfo: SIG#-4194561
21268/1:        lwp_sigtimedwait(0xFFBFF1A0, 0xFFBFF228, 0x00000010) = 0
21268/1:                sigmask = 0xFFBFFEFF 0x0000FF1F 0 0
21268/1:              siginfo: SIG#-4194561
21268/1:        sigprocmask(3, 0xFFBFF228, 0xFFBFF238)          = 0
21268/1:                 set = 0xFFBFFEFF 0x0000FF1F 0 0
21268/1:                oset = 0x00002000 0 0 0
21268/1:        lwp_sigtimedwait(0xFFBFF238, 0xFFBFF1B0, 0x00000010) = 0
21268/1:                sigmask = 0x00002000 0 0 0
21268/1:              siginfo: SIG#8192
21268/1:        lwp_sigtimedwait(0xFFBFF1A0, 0xFFBFF360, 0x00000010) = 0
21268/1:                sigmask = 0x00002000 0 0 0
21268/1:              siginfo: SIG#8192
21268/1:        lwp_park(1, 1, 1)                               = 0
21268/1:        lwp_sigtimedwait(0xFFBFF370, 0xFFBFF088, 0x000001C0) = 0
21268/1:                sigmask = 0x0000082F 0 0 0
21268/1:              siginfo: SIG#2095
21268/1:        lwp_sigtimedwait(0xFFBFF090, 0xFFBFEFF8, 0x00000010) = 0
21268/1:                sigmask = 0 0 0 0
21268/1:              siginfo: SIG#0
21268/1:        lwp_sigtimedwait(0xFFBFEFE8, 0xFFBFF090, 0x00000010) = 0
21268/1:                sigmask = 0 0 0 0
21268/1:              siginfo: SIG#0
21268/1:        context(1, 0xFFBFF088)
21268/1:        getpid()                                        = 21268 [1417]
21268/1:        kill(1417, 0x0000000F)                          = 0
21268/1:        alarm(1)                                        = 0
21268/1:            Received signal #14, SIGALRM, in pause() [caught]
21268/1:              siginfo: SIG#0
21268/1:        pause()                                         Err#4 EINTR

No related posts.

This post is password protected. To view it please enter your password below:

Password:

No related posts.

What Prowl does is send “push” events to your iPhone from virtually anything.
Install the app, register on the Prowl website, generate an API key, and you’re home and hosed for some good hacking.

This means that any application that wants to push a notification to your iPhone contacts the Prowl servers, which then sends the notification on to your iPhone. So far it’s reliable enough, but even they say not to trust it for critical or emergency applications.

There is an API, and it is fairly complete, supporting all your language preferences, (Java, perl, PHP, etc, etc – sorry no COBOL).

Of course, there’s even a plugin for WordPress, (which I have to say is so much easier than Joomla). It will send push alerts for new posts, comments, and other things.

WordPress post alert

Looks like someone has posted another blog entry.

Viewing the notification brings you here:

Alerts

Alerts can be adjusted, the nice feature is being able to silence them for a given time.

And adjust what sounds are played and when, although the default sounds are too quiet I find.

Redirects

Redirects allow a specified application to start when a particular event comes in. Trouble is that there’s no way of selecting anything else but the given set of applications. Maybe in the next release.

No related posts.

This was for several reasons:
1. Joomla is a pain in the bum for blogging. Its too complex for a blog and I keep forgetting the differences between a “com”, “module” and a “plugin”.
2. Upgrading is a pain in the bum. I just recently noticed that half my content went missing as part of a recent upgrade.
3. I had bought an iPhone and wanted to be able to post while out and about. Getting this going in Joomla was a pain in the bum, (although I discovered after switching to WordPress that it wasn’t really Joomla’s fault).
4. My old TreeCam blog was running on an old version of s9y, and I wanted to combine it all.

Not that Joomla is all bad, it’s a great CMS app, but just not for blogging. It has a fantastic plugin, (module or whatever the heck you call it), for eCommerce, (virtuemart), and even a blog thingy. I always found it to be a struggle. One of those things. It’s not designed for blogging.

It was a good decision to move. WordPress is infinitely easier than Joomla. None of the learning curve. So far there seems to be the same number of plugins for both, but it looks like Joomla is getting a little “stale”. Installing plugins for WordPress is damn easy. Almost a “one click” install or upgrade for all plugins and templates.

Since the switch I’ve discovered all sorts of goodies. EG out of the box weblog API, RPC calls, blog pings, dead link monitoring.

Nice. Really polished app.

Even better is the fact that I didn’t need to do anything for my iPhone to start posting to my blog. Just install BlogPress, (and the free WordPress app), from the Apple app store, enter in my site, and start blogging. I even used it to migrate from Joomla to WordPress, as you can post to multiple sites in one hit.

BlogPress’ settings page

You can post piccies and videos to Picasa and YouTube. Specify multiple blogs to post to.

Pulling down posts

Posting a story

Is dead easy, although may be a problem for those who don’t know their HTML.

Inserting images

Just click on the camera icon top right, and you can insert photos directly into the post.

Save or post?

You can post it, or save it as a draft for later.

Posting an entry

You’re also given the option of viewing it in Safari, which is a minor annoyance for me. It’d be good to be able to turn it off.

– Posted from my iPhone

No related posts.

Sun’s Java Web Console is a nice little toy. Great for those people who prefer the less extreme learning curve of a CLI, and great for manager types who like pictures.

I’ve used it several times before, but thought I’d try it in on x86 Solaris10u6 box I have here. Bloody Sun. You think a released version implies that it’s stable, and working.

For me the issue was that when I originally installed the server, I used a temporary hostname, and DHCP for network because I couldn’t be bothered to change my local DNS and figure out a spare IP. I then changed to fixed IP and hostname later.

The ramifications of this is that my hostname ended up being assigned to localhost, (aka 127.0.0.1). This means that when the WebConsole was initially installed and configured it used 127.0.0.1 throughout it’s hardcoded, (stupid Tomcat – hate it with a vengence), config files.

This is something that annoys the heck out of me: I understand developers making ‘packages’ that they can ‘deploy’ to a web server, (Tomcat). It’s nice to be able to ‘deploy’ a package. But for heck’s sake, why shove hostname/IP/port definitions in not just one file, but 50 stupid files that are zipped up along with shared libs and binaries? Why? Haven’t they heard of /etc/hosts, or even putting these definitions into somewhere common? Why do they have to have these definitions in multiple places?

Another great example of letting the developer take the lead in systems management.

Anyway, digging around the net trying to resolve my issue showed up a lot of similar complaints, but nothing really to sort it out. ‘wcadmin’ refused to start the ‘deployed’ packages, because they couldn’t connect to the Tomcat server running on my old hostname.

root@somewhere % wcadmin list

Deployed web applications (application name, context name, status):

console  ROOT            [stopped]
console  com_sun_web_ui  [stopped]
console  console         [stopped]
console  manager         [stopped]
console  zfs             [stopped]

Registered jar files (application name, identifier, path):

console  audit_jar     /usr/lib/audit/Audit.jar
console  console_jars  /usr/share/webconsole/lib/*.jar
console  jato_jar      /usr/share/lib/jato/jato.jar
console  javahelp_jar  /usr/jdk/packages/javax.help-2.0/lib/*.jar
console  shared_jars   /usr/share/webconsole/private/container/shared/lib/*.jar

Registered login modules (application name, service name, identifier):

console  ConsoleLogin  userlogin
console  ConsoleLogin  rolelogin

Persistent Jvm options:

-server
-Xmx128m
-XX:+UseParallelGC
-XX:ParallelGCThreads=4

I tried using wcadmin ro ‘undeploy’ and ‘deploy’ packages, and even the brut force method of

root@somewhere % find /var/webconsole /usr/share/webconsole /etc/webconsole /usr/lib/webconsole -type f | xargs egrep -l '127.0.0.1|localhost'

and manually modifying any config files found.

In the end I did a pkgrm/remove configs/pkgadd cycle:

1. Remove WebConsole packages.

system      SUNWasac  Sun Java System Application Server Admin Client
application SUNWmcon  Sun Java(TM) Web Console 3.0.2 (Core)
system      SUNWmconr Sun Java(TM) Web Console 3.0.2 (Root)
application SUNWmcos  Implementation of Sun Java(TM) Web Console (3.0.2) services
application SUNWmcosx Implementation of Sun Java(TM) Web Console (3.0.2) services
application SUNWmctag Sun Java(TM) Web Console 3.0.2 (Tags & Components)
application SUNWzfsgr ZFS Administration for Sun Java(TM) Web Console (Root)
application SUNWzfsgu ZFS Administration for Sun Java(TM) Web Console (usr)

The usual removal:

root@somewhere % pkgrm SUNWzfsgu SUNWmcosx SUNWmcos SUNWasac SUNWmcon SUNWzfsgr SUNWmconr SUNWmctag

Then remove all the config files:

root@somwehere % rm -rf /var/webconsole /usr/share/webconsole /etc/webconsole /usr/lib/webconsole

Or tarball up and then remove:

root@somewhere % tar cvf /tmp/webconsole.tar /var/webconsole /usr/share/webconsole /etc/webconsole /usr/lib/webconsole
root@somewhere % rm -rf /var/webconsole /usr/share/webconsole /etc/webconsole /usr/lib/webconsole

2. Re-install the WebConsole packages:
If you have your original install DVD/CD handy, then just chuck it in, and install the packages from the Product directory:

root@somewhere % pkgadd -d /cdrom/cdrom0/s2/Solaris_10/Product SUNWzfsgu SUNWmcosx SUNWmcos SUNWasac SUNWmcon SUNWzfsgr SUNWmconr SUNWmctag
The following packages are available:
 1  SUNWasac      Sun Java System Application Server Admin Client
 (i386) 8.2,REV=2006.08.16.08.46
 2  SUNWmcon      Sun Java(TM) Web Console 3.0.2 (Core)
 (i386) 3.0.2,REV=2006.12.08.20.48
 3  SUNWmconr     Sun Java(TM) Web Console 3.0.2 (Root)
 (i386) 3.0.2,REV=2006.12.08.23.39
 4  SUNWmcos      Implementation of Sun Java(TM) Web Console (3.0.2) services
 (i386) 3.0.2,REV=2006.12.08.23.39
 5  SUNWmcosx     Implementation of Sun Java(TM) Web Console (3.0.2) services
 (i386) 3.0.2,REV=2006.12.08.23.39
 6  SUNWmctag     Sun Java(TM) Web Console 3.0.2 (Tags & Components)
 (i386) 3.0.2,REV=2006.12.08.20.48
 7  SUNWzfsgr     ZFS Administration for Sun Java(TM) Web Console (Root)
 (i386) 1.0,REV=2006.10.24.20.44
 8  SUNWzfsgu     ZFS Administration for Sun Java(TM) Web Console (usr)
 (i386) 1.0,REV=2006.10.24.20.44

Even though you are installing version 3.1 of the Java Web Console, it actually appears as 3.0.2 for some stupid reason.

Everything will startup as expected using the new hostname defined.

root@somewhere % wcadmin list

Deployed web applications (application name, context name, status):

console  ROOT            [running]
console  com_sun_web_ui  [running]
console  console         [running]
console  manager         [running]
console  zfs             [running]

Registered jar files (application name, identifier, path):

console  audit_jar     /usr/lib/audit/Audit.jar
console  console_jars  /usr/share/webconsole/lib/*.jar
console  jato_jar      /usr/share/lib/jato/jato.jar
console  javahelp_jar  /usr/jdk/packages/javax.help-2.0/lib/*.jar
console  shared_jars   /usr/share/webconsole/private/container/shared/lib/*.jar

Registered login modules (application name, service name, identifier):

console  ConsoleLogin  userlogin
console  ConsoleLogin  rolelogin

Persistent Jvm options:

-server
-Xmx128m
-XX:+UseParallelGC
-XX:ParallelGCThreads=4

No related posts.

No related posts.

% pkg install gnome

Or

% pkg remove hugin

Or

% pkg verify firefox

Or

% pkg list

I mean…. Why not Sun?

(And what about my cat?)

No related posts.

So, this howto shows you how to run Opera Mini on Linux in glorious 1440×900…. and take advantage of the speed increase.

Requirements

The Linux distribution I run is Debian, but this HOWTO will apply to any distribution really. In fact it should work on Windows as well. All you need initially is a running version of Java. However the complete list is:
Java 6
Micro Emulator
Opera Mini

Installing Java

On Debian I just do the usual installation process:

root@localhost % apt-get install sun-java6-bin sun-java6-jre

That’s it for java.

Installing MicroEmulator

Now for MicroEmulator. Fetch whatever the latest release is . I used 2.0.2 for this setup.

Extract the tarball to somewhere. Doesn’t matter where. I put it under $HOME/lib

mick@localhost [~/lib] # tar zxvf microemulator-2.0.2.tar.gz

I put in a symlink so that I can reference different version if need be. But you can skip this step if you want.

mick@localhost [~/lib] # ln -s microemulator-2.0.2 microemulator

Now, fire up java and point it to the microemulator jar file.

mick@localhost [~/lib] # java -jar $HOME/lib/microemulator/microemulator.jar

This window will now pop-up.

Now Click on ‘Options’, and then ‘Select device’. There’s nothing stopping you from running Opera in this virtual device. But remember you want to run it in a big window, right!

The window below will pop-up.

Click on Add… You want to traverse to your {directory where you installed the microemulator}/devices/, and select the microemu-device-resizable.jar file.

It will now appear in your virtual devices list. Click on ‘Set as default’ to use it every time you start it up.

The MicroEmulator window will change to a ‘not as snazzy’ window like this:

That’s it for the microemulator.

Installing Opera Mini

Installing Opera Mini is easy too. There’s two ways to do it.
Click here , and here and save somewhere.

Or follow these instructions, (I’ve put them here, because, as you know, links change on the internet, and I hate 404s).

Just fetch the latest version from Opera. There’s 4 steps to download:
1. Select ‘Download to PC’.
2. Select ‘Generic version’.
3. Select ‘Advanced MIDP 2′
4. Select you preferred language, and download the JAD and JAR files.

I placed these two files under $HOME/lib/opera. You will have to create a symlink so that the version you have downloaded will be used properly.

mick@localhost [~/lib/opera] # ln -s opera-mini-latest-advanced-int.jar opera-mini-4.1.11355-advanced-int.jar

Your version may/will vary. You’ll need to symlink, (or rename), the file for it to work, though.

Now you want to open the Opera Mini JAD file you just downloaded.

Find it and open it.

Now you will see it in the list of applications that can be launched by Micro Emulator.

Now just double click on it.

Hooley Dooley Batman! It works! Cool!

Either use your mouse and click, or you can use up/down/left/right keys and enter to traverse web pages.

One last thing

I created the following script to make it easier for me to startup, and called it ‘operamini’.

#!/bin/bash

exec >& /tmp/operamini.log

java -jar $HOME/lib/microemulator/microemulator.jar $HOME/lib/opera/opera-mini-latest-advanced-int.jad

Now I have my super fast GPRS internet connection. Also, a side benefit is that all my bookmarks are synced up to my phone.

No related posts.