Search results for query: *

Switching from Link Based to Probe Base IPMP uses three IP's instead of one which is quite an overhead. Is a mixture of 1 x IP and 2 x Private IP's the answer? My proposed setup would be : hostname.bge0 host-bge0 netmask + broadcast + deprecated -failover group IPMP-1 up \ addif host1 netmask...

In Solaris10 I changed NIC drivers from bge0 to bcme0 via package install with no problems. I disassociated bge and after a reboot and some hostname changes new driver loaded OK. Now trying to rollback after unplumb and downing bcme0 I cannot plumb bge0 ifconfig bge0 plumb ifconfig: cannot open...

killing slapd is no good as it restarts itself immediately. Any advice please? Thanks Marrow

Solaris10 /Oracle 10, Oracle Clustering. Nothing reported in system messages or fmadm, nothing in alerts, nothing in rac files, just crashing sometimes side1 or 1 and lately 1 & 2. Memory / CPU good but I/O high on some SAN disks, cannot find any clues.

iostat -En on Solaris9 would display disk in descriptive mode cXtYdZ but this appears not to work in S10. I am trying to equate an ssd to an actual disk. Any ideas please as to why this no longer works?

I have displayed & tabbed 4 important columns of output from "iostat -xtc 2 10" as follows iostat -xtc 2 10 | awk '{printf("%\t%s\t%s\t%s\t%s\n",$1,$2,$3,$8)}' 1) I would now like to discard all lines that show 0.0 0.0 0.0 0.0 2) or, only display anything greater than 30 in any column...

Need some information on 3.2 please. Is it a major difference to 3.1, does it have it's own "new" commands and do the above no longer function?

I want to run "iostat -xtc 2 5 and delete first set of "extended" stats and print 2nd 3rd 4th and 5th set using SED.

On Solaris8 & veritas I have a (ufs)f/s called /u01/mgmt/data that uses 48GB on disk2 but f/s is 80% full and disk has no spare capacity. I have a new empty 72Gb disk3 and would like to add another 48Gb for my f/s to make 96Gb in total over the two disks. Best way to do this?

Mar 2 06:26:58 hostname Mar 2 06:26:58 hostname scsi: [ID 107833 kern.warning] WARNING: /pci@8,700000/ide@1/sd@0,0 (sd0): Mar 2 06:26:58 hostname transport rejected bad packet Any Cause for concern, happens every 10mins and the last month. Network isn't that stressed..... hostname...

Used OK on a 2Gb memory XP laptop, but on a desktop with only 512Mb get the above. For what it's worth it will boot in Failsafe/ single user. Is this memory shortage or wrong version wrong system?

I have some test systems, Solaris9>10, can I install veritas without disk arrays and simply mirror as in SDS? If so which s/w has to be downloaded. Thanks for ay advice

Orace9.2 SQL> connect /as sysdba ERROR: ORA-09817: Write to audit file failed. SVR4 Error: 28: No space left on device ORA-09945: Unable to initialize the audit trail file SVR4 Error: 28: No space left on device On test system USERS.dbf has taken 12Gb of space and filled file-system to 100%...

value=25 If [ value = 25 ] then... # this works OK But also in ksh If [ value -gt 24 ] then.... # does not work. Any advice please? Is -lt or -gt not vaid in ksh?

Have found a test system with 2 mirrored disks but only 2 x metadb's. Would like to increase to 4 x metadb's but I believe you have to delete originals first. Also do I have to clear mirrors or just modify /etc/system? Thanks in advance

Anybody used Solaris10 dtrace and any simple examples please?

Have an old test E450 not mirrored (Solaris8) 3 x 9Gb Disks. /u01 is full of data on say disk2. Have spare empty 9Gb on disk3, need to add new space to u01. What's the best method of doing this, just a link?

I want to find every line with... "Receive Transaction" in "myfile" then ignore this line and the next 11 lines then select the next 2 lines.... 'Rcv-0' 'ROUTER_x' 'Rcv-1' 'ROUTER_y' join these 2 lines together to make.... 'Rcv-0' 'ROUTER_x''Rcv-1' 'ROUTER_y' Then repeat through the file I...

In oracle9, I have compiled oracle procedure "CREATE OR REPLACE EXTRACT_FILE1 AS ...." I can run OK if sqlplus /nolog >connect user1/user1@database1 >execute EXTRACT_FILE1 But not sure how to use short form... Can I initiate the execute from unix on single line? sqlplus...

Have a pair of clustered E450 3Gb Memory, 4Gb swap, have 170 processes each and 2 & 1 ora-dbases per side . Reboot clears, but lose 800Mb immediately on startup. Our DBA likes running daily oracle processes into /TMP (1.8Mb of their data always there), we don't like this at all but nothing we...