SGD 4 serious problems.
From the forum you can see I have had many issues, of which I believed were
all fixed when I removed Tarantella from my server and re-installed it 3
days ago. With an average of 2 to 8 users logged in over those 3 days
something still managed to go wrong. Is their something very wrong with SGD
4 that users need to know about? This is the worst fresh install of nay
software I have ever seen in my years in I.T. In the last week I have come
across one problem after another and it doesn't seem to be getting any
better. Last night the whole server crashed and all I got was the errors
below from the script I run to get the number of logged in users (tarantella
license query). This is on Solaris 9 w/ dual Sparc 450 Mhz and 2gigs of RAM
of which 150 Mb is reserved for SGD. As I said this issue cropped up with 3
days of uptime on a dedicated SGD server recently built.
Error occurred during initialization of VM
Could not reserve enough space for object heap
Error occurred during initialization of VM
Could not reserve enough space for code cache
Error occurred during initialization of VM
Could not reserve enough space for code cache
Error occurred during initialization of VM
Could not reserve enough space for code cache
Error occurred during initialization of VM
Could not reserve enough space for code cache
Error occurred during initialization of VM
Could not reserve enough space for object heap
Exception java.lang.OutOfMemoryError: requested 80044 bytes for
symbolTableBucket in /exp
ort1/jdk142-update/ws/fcs/hotspot/src/share/vm/memory/symbolTable.cpp. Out
of swap space?
You must be root to run this program
Error occurred during initialization of VM
Could not reserve enough space for object heap
Error occurred during initialization of VM
Could not reserve enough space for object heap
Exception java.lang.OutOfMemoryError: requested 80044 bytes for
symbolTableBucket in /exp
ort1/jdk142-update/ws/fcs/hotspot/src/share/vm/memory/symbolTable.cpp. Out
of swap space?
From my system monitor all i could about the system was this:
PIDUIDLWP PRI NICE SIZERES STATETIMECPU COMMAND
2621001136 2910 123M84M sleep 127:17 34.48% java
18911014 291082M41M sleep 128:26 32.35% java
2145701301077M37M run95:40 28.30% java
[2504 byte] By [
Peter] at [2007-11-25 20:52:49]
# 1
Are their any issues with Solaris 9 and the version of Java that SGD 4 is
using? I had this same issue again this morning, at least I was able to get
into the system and do a tarantella restart --warm. This fixed the problem
but now drive mapping doesn't work. Does anyone else see this issue where
the process consumes the CPU after a certain period of time?
"Peter" <gioti1973@yahoo.com> wrote in message
news:d6v4fn$tkf$1@mint.tarantella.com...
> From the forum you can see I have had many issues, of which I believed
were
> all fixed when I removed Tarantella from my server and re-installed it 3
> days ago. With an average of 2 to 8 users logged in over those 3 days
> something still managed to go wrong. Is their something very wrong with
SGD
> 4 that users need to know about? This is the worst fresh install of nay
> software I have ever seen in my years in I.T. In the last week I have come
> across one problem after another and it doesn't seem to be getting any
> better. Last night the whole server crashed and all I got was the errors
> below from the script I run to get the number of logged in users
(tarantella
> license query). This is on Solaris 9 w/ dual Sparc 450 Mhz and 2gigs of
RAM
> of which 150 Mb is reserved for SGD. As I said this issue cropped up with
3
> days of uptime on a dedicated SGD server recently built.
>
> Error occurred during initialization of VM
> Could not reserve enough space for object heap
> Error occurred during initialization of VM
> Could not reserve enough space for code cache
> Error occurred during initialization of VM
> Could not reserve enough space for code cache
> Error occurred during initialization of VM
> Could not reserve enough space for code cache
> Error occurred during initialization of VM
> Could not reserve enough space for code cache
> Error occurred during initialization of VM
> Could not reserve enough space for object heap
>
> Exception java.lang.OutOfMemoryError: requested 80044 bytes for
> symbolTableBucket in /exp
> ort1/jdk142-update/ws/fcs/hotspot/src/share/vm/memory/symbolTable.cpp. Out
> of swap space?
> You must be root to run this program
> Error occurred during initialization of VM
> Could not reserve enough space for object heap
> Error occurred during initialization of VM
> Could not reserve enough space for object heap
>
> Exception java.lang.OutOfMemoryError: requested 80044 bytes for
> symbolTableBucket in /exp
> ort1/jdk142-update/ws/fcs/hotspot/src/share/vm/memory/symbolTable.cpp. Out
> of swap space?
>
> From my system monitor all i could about the system was this:
>
> PIDUIDLWP PRI NICE SIZERES STATETIMECPU COMMAND
>2621001136 2910 123M84M sleep 127:17 34.48% java
> 18911014 291082M41M sleep 128:26 32.35% java
> 2145701301077M37M run95:40 28.30% java
>
>
>
Peter at 2007-7-4 19:01:07 >
# 2
After a little more investigation it seems that there are two java processes
that get spawned by root.....?...which is what causes the cpu overflow.
Why on earth is root spawning these two java processes? and why only after a
certain period of time? Anyone else seen this?ttasys is the sgd user but
when the java problem crops up I see that this using top:
last pid: 26131; load averages: 2.46, 2.42, 2.36
16:46:29
139 processes: 136 sleeping, 1 running, 2 on cpu
CPU states: 1.3% idle, 85.6% user, 13.1% kernel, 0.0% iowait, 0.0% swap
Memory: 2048M real, 245M free, 1612M swap in use, 1416M swap free
PID USERNAME LWP PRI NICE SIZERES STATETIMECPU COMMAND
262 ttasys154010 121M77M run50:12 34.68% java
22969 root13 291076M27M sleep25:11 27.86% java
20986 root1301077M27M cpu/243:11 27.73% java
"Peter" <gioti1973@yahoo.com> wrote in message
news:d71vhe$bqf$1@mint.tarantella.com...
> Are their any issues with Solaris 9 and the version of Java that SGD 4 is
> using? I had this same issue again this morning, at least I was able to
get
> into the system and do a tarantella restart --warm. This fixed the problem
> but now drive mapping doesn't work. Does anyone else see this issue where
> the process consumes the CPU after a certain period of time?
>
>
> "Peter" <gioti1973@yahoo.com> wrote in message
> news:d6v4fn$tkf$1@mint.tarantella.com...
> > From the forum you can see I have had many issues, of which I believed
> were
> > all fixed when I removed Tarantella from my server and re-installed it 3
> > days ago. With an average of 2 to 8 users logged in over those 3 days
> > something still managed to go wrong. Is their something very wrong with
> SGD
> > 4 that users need to know about? This is the worst fresh install of nay
> > software I have ever seen in my years in I.T. In the last week I have
come
> > across one problem after another and it doesn't seem to be getting any
> > better. Last night the whole server crashed and all I got was the errors
> > below from the script I run to get the number of logged in users
> (tarantella
> > license query). This is on Solaris 9 w/ dual Sparc 450 Mhz and 2gigs of
> RAM
> > of which 150 Mb is reserved for SGD. As I said this issue cropped up
with
> 3
> > days of uptime on a dedicated SGD server recently built.
> >
> > Error occurred during initialization of VM
> > Could not reserve enough space for object heap
> > Error occurred during initialization of VM
> > Could not reserve enough space for code cache
> > Error occurred during initialization of VM
> > Could not reserve enough space for code cache
> > Error occurred during initialization of VM
> > Could not reserve enough space for code cache
> > Error occurred during initialization of VM
> > Could not reserve enough space for code cache
> > Error occurred during initialization of VM
> > Could not reserve enough space for object heap
> >
> > Exception java.lang.OutOfMemoryError: requested 80044 bytes for
> > symbolTableBucket in /exp
> > ort1/jdk142-update/ws/fcs/hotspot/src/share/vm/memory/symbolTable.cpp.
Out
> > of swap space?
> > You must be root to run this program
> > Error occurred during initialization of VM
> > Could not reserve enough space for object heap
> > Error occurred during initialization of VM
> > Could not reserve enough space for object heap
> >
> > Exception java.lang.OutOfMemoryError: requested 80044 bytes for
> > symbolTableBucket in /exp
> > ort1/jdk142-update/ws/fcs/hotspot/src/share/vm/memory/symbolTable.cpp.
Out
> > of swap space?
> >
> > From my system monitor all i could about the system was this:
> >
> > PIDUIDLWP PRI NICE SIZERES STATETIMECPU COMMAND
> >2621001136 2910 123M84M sleep 127:17 34.48% java
> > 18911014 291082M41M sleep 128:26 32.35% java
> > 2145701301077M37M run95:40 28.30% java
> >
> >
> >
>
>
Peter at 2007-7-4 19:01:07 >
