Oninit cdi / Archive
Talk to an expert →

chunks status is offline

Posted in 2007

Summary

Gulshan Kumar, new to Informix 9.30 on Solaris, had mirrored chunks showing PD and MD status. Art Kagel suggested dropping the damaged chunk and re-adding a mirror. Marcus Haarmann suggested onspaces -s space -p path -o 0 -O, which failed to open the file or size. He advised checking permissions, space and metastat, since the Solaris metadevice may be down. Unresolved.

Auto-generated by Claude from the posts below — may be imperfect; read the full thread.

Topics: Storage & Space Management, Platform-Specific Issues

GULSHAN KUMAR — — source: IIUG Forums & Mailing Lists

I am using IDS.9.30.UC2 on solaris. The below is the output of onstat -d command. Dbspaces address number flags fchunk nchunks flags owner name 121007d0 1 0x2 1 1 M informix rootdbs 1216eec0 2 0x2 2 1 M informix dbs01 1215fa40 3 0x2 3 1 M informix dbs02 1215fb88 4 0x2 4 1 M informix dbs03 1215fcd0 5 0x2 5 1 M informix dbs04 1215fe18 6 0x2 6 1 M informix dbs05 127d5018 7 0x1 7 1 N informix temp1 127d5160 8 0x8001 8 1 N S informix sbdbs01 8 active, 2047 maximum Chunks address chk/dbs offset size free bpages flags pathname 12100918 1 1 0 1048000 162683 PO- /informix/ol_insrvdb05/root 12100a88 1 1 0 1048000 0 MO- /informix/ol_insrvdb05/rootm 1216ea70 2 2 0 1048575 82918 PO- /informix/ol_insrvdb05/dbs01 1215f310 2 2 0 1048575 0 MD- /informix/ol_insrvdb05/dbs01m 1216ebe0 3 3 0 1048575 0 PD- /informix/ol_insrvdb05/dbs02 1215f480 3 3 0 1048575 286243 MO- /informix/ol_insrvdb05/dbs02m 1216ed50 4 4 0 1048575 228124 PO- /informix/ol_insrvdb05/dbs03 1215f5f0 4 4 0 1048575 0 MD- /informix/ol_insrvdb05/dbs03m 12100c30 5 5 0 1048575 0 PD- /informix/ol_insrvdb05/dbs04 1215f760 5 5 0 1048575 3383 MO- /informix/ol_insrvdb05/dbs04m 12100da0 6 6 0 1048575 450661 PO- /informix/ol_insrvdb05/dbs05 1215f8d0 6 6 0 1048575 0 MO- /informix/ol_insrvdb05/dbs05m 1215f030 7 7 0 1048575 1015772 PO- /informix/ol_insrvdb05/temp1 1215f1a0 8 8 0 256000 238979 238979 POS /export/home/sbdbs01 Metadata 16968 12832 16968 In the above output it is showing the status of few chunks PD / MD. How can i change the stautus of these chunks. How to recover the corrupted chunks from mirror or vice versa. Please suggest as i am new to informix and don't know how to recover. Thanks in Advance. Gulshan

↪ replying to GULSHAN KUMAR
ART KAGEL, BLOOMBERG/ 731 LEXIN — — source: IIUG Forums & Mailing Lists

All you have to do is to drop the damaged chunk and readd a new mirror for it from a good disk structure. Then engine will automatically recover it. Art S. Kagel ----- Original Message ----- From: Gulshan Kumar <ids@iiug.org> To: ids@iiug.org At: 10/15 9:49:56 I am using IDS.9.30.UC2 on solaris. The below is the output of onstat -d command. Dbspaces address number flags fchunk nchunks flags owner name 121007d0 1 0x2 1 1 M informix rootdbs 1216eec0 2 0x2 2 1 M informix dbs01 1215fa40 3 0x2 3 1 M informix dbs02 1215fb88 4 0x2 4 1 M informix dbs03 1215fcd0 5 0x2 5 1 M informix dbs04 1215fe18 6 0x2 6 1 M informix dbs05 127d5018 7 0x1 7 1 N informix temp1 127d5160 8 0x8001 8 1 N S informix sbdbs01 8 active, 2047 maximum Chunks address chk/dbs offset size free bpages flags pathname 12100918 1 1 0 1048000 162683 PO- /informix/ol_insrvdb05/root 12100a88 1 1 0 1048000 0 MO- /informix/ol_insrvdb05/rootm 1216ea70 2 2 0 1048575 82918 PO- /informix/ol_insrvdb05/dbs01 1215f310 2 2 0 1048575 0 MD- /informix/ol_insrvdb05/dbs01m 1216ebe0 3 3 0 1048575 0 PD- /informix/ol_insrvdb05/dbs02 1215f480 3 3 0 1048575 286243 MO- /informix/ol_insrvdb05/dbs02m 1216ed50 4 4 0 1048575 228124 PO- /informix/ol_insrvdb05/dbs03 1215f5f0 4 4 0 1048575 0 MD- /informix/ol_insrvdb05/dbs03m 12100c30 5 5 0 1048575 0 PD- /informix/ol_insrvdb05/dbs04 1215f760 5 5 0 1048575 3383 MO- /informix/ol_insrvdb05/dbs04m 12100da0 6 6 0 1048575 450661 PO- /informix/ol_insrvdb05/dbs05 1215f8d0 6 6 0 1048575 0 MO- /informix/ol_insrvdb05/dbs05m 1215f030 7 7 0 1048575 1015772 PO- /informix/ol_insrvdb05/temp1 1215f1a0 8 8 0 256000 238979 238979 POS /export/home/sbdbs01 Metadata 16968 12832 16968 In the above output it is showing the status of few chunks PD / MD. How can i change the stautus of these chunks. How to recover the corrupted chunks from mirror or vice versa. Please suggest as i am new to informix and don't know how to recover. Thanks in Advance. Gulshan ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.

↪ replying to GULSHAN KUMAR
Marcus Haarmann — — source: IIUG Forums & Mailing Lists

Hi, First you have to see that the devices are available again with enough space and with correct permissions (660 informix:informix). Then you can use the onspaces utility to bring a chunk back to online state. Syntax is onspaces -s <spacename> -p <path> -o <offset> {-O | -D} [-y] } In your case (example for dbs01): onspaces -s dbs01 -p /informix/ol_insrvdb05/dbs01m -o 0 -O Marcus -----Ursprüngliche Nachricht----- Von: GULSHAN KUMAR [mailto:gulshank@infotechsw.com] Gesendet: Montag, 15. Oktober 2007 15:49 An: ids@iiug.org Betreff: chunks status is offline [10136] I am using IDS.9.30.UC2 on solaris. The below is the output of onstat -d command. Dbspaces address number flags fchunk nchunks flags owner name 121007d0 1 0x2 1 1 M informix rootdbs 1216eec0 2 0x2 2 1 M informix dbs01 1215fa40 3 0x2 3 1 M informix dbs02 1215fb88 4 0x2 4 1 M informix dbs03 1215fcd0 5 0x2 5 1 M informix dbs04 1215fe18 6 0x2 6 1 M informix dbs05 127d5018 7 0x1 7 1 N informix temp1 127d5160 8 0x8001 8 1 N S informix sbdbs01 8 active, 2047 maximum Chunks address chk/dbs offset size free bpages flags pathname 12100918 1 1 0 1048000 162683 PO- /informix/ol_insrvdb05/root 12100a88 1 1 0 1048000 0 MO- /informix/ol_insrvdb05/rootm 1216ea70 2 2 0 1048575 82918 PO- /informix/ol_insrvdb05/dbs01 1215f310 2 2 0 1048575 0 MD- /informix/ol_insrvdb05/dbs01m 1216ebe0 3 3 0 1048575 0 PD- /informix/ol_insrvdb05/dbs02 1215f480 3 3 0 1048575 286243 MO- /informix/ol_insrvdb05/dbs02m 1216ed50 4 4 0 1048575 228124 PO- /informix/ol_insrvdb05/dbs03 1215f5f0 4 4 0 1048575 0 MD- /informix/ol_insrvdb05/dbs03m 12100c30 5 5 0 1048575 0 PD- /informix/ol_insrvdb05/dbs04 1215f760 5 5 0 1048575 3383 MO- /informix/ol_insrvdb05/dbs04m 12100da0 6 6 0 1048575 450661 PO- /informix/ol_insrvdb05/dbs05 1215f8d0 6 6 0 1048575 0 MO- /informix/ol_insrvdb05/dbs05m 1215f030 7 7 0 1048575 1015772 PO- /informix/ol_insrvdb05/temp1 1215f1a0 8 8 0 256000 238979 238979 POS /export/home/sbdbs01 Metadata 16968 12832 16968 In the above output it is showing the status of few chunks PD / MD. How can i change the stautus of these chunks. How to recover the corrupted chunks from mirror or vice versa. Please suggest as i am new to informix and don't know how to recover. Thanks in Advance. Gulshan ************************************************************************ ******* Forum Note: Use "Reply" to post a response in the discussion forum.

↪ replying to Marcus Haarmann
GULSHAN KUMAR — — source: IIUG Forums & Mailing Lists

Hi Marcus, Thanks for your valuable reply. Here is the output of below command: bash-2.03$ onspaces -s dbs01 -p /informix/ol_insrvdb05/dbs01m -o 0 -OWARNING: bringing a down chunk into recovery state. Do you really want to continue? (y/n)y Verifying physical disk space, please wait ... Error opening file /informix/ol_insrvdb05/dbs01m. bash-2.03$ onspaces -s dbs02 -p /informix/ol_insrvdb05/dbs02 -o 0 -OWARNING: bringing a down chunk into recovery state. Do you really want to continue? (y/n)y Verifying physical disk space, please wait ... The chunk '/informix/ol_insrvdb05/dbs02' will not fit in the space specified. The chunks are created on raw devices: lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs01 -> /dev/md/rdsk/d12 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs01m -> /dev/md/rdsk/d20 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs02 -> /dev/md/rdsk/d21 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs02m -> /dev/md/rdsk/d41 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs03 -> /dev/md/rdsk/d31 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs03m -> /dev/md/rdsk/d61 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs04 -> /dev/md/rdsk/d22 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs04m -> /dev/md/rdsk/d32 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs05 -> /dev/md/rdsk/d52 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs05m -> /dev/md/rdsk/d42 lrwxrwxrwx 1 root other 16 Dec 13 2005 er_dbs1 -> /dev/md/rdsk/d40 lrwxrwxrwx 1 root other 16 Dec 13 2005 er_dbs1m -> /dev/md/rdsk/d51 lrwxrwxrwx 1 root other 16 Dec 13 2005 er_sbdbs1 -> /dev/md/rdsk/d50 lrwxrwxrwx 1 root other 16 Dec 13 2005 er_sbdbs1m -> /dev/md/rdsk/d60 lrwxrwxrwx 1 root other 16 Dec 13 2005 root -> /dev/md/rdsk/d10 lrwxrwxrwx 1 root other 16 Dec 13 2005 rootm -> /dev/md/rdsk/d30 lrwxrwxrwx 1 root other 16 Dec 13 2005 temp1 -> /dev/md/rdsk/d11 Now kindly suggest how to check the permissions and how to proceed further.

↪ replying to GULSHAN KUMAR
Marcus Haarmann — — source: IIUG Forums & Mailing Lists

Hi, That is what I thought. You have permissions problem or capacity problem. Check where the dbs01m file points to and modify the permissions (chown informix:informix dbs01m; chmod 660 dbs01m). Make sure the location is not used anywhere else and gives enough room to store your data. As I can see, the device is pointing to a md group. On solaris, use the metastat command to verify the status of the devices. I believe, it is a stripe setup, which is down due to a failed disk. When you have cleared up the meta devices, you can bring up the chunks again. Marcus -----Ursprüngliche Nachricht----- Von: GULSHAN KUMAR [mailto:gulshank@infotechsw.com] Gesendet: Montag, 15. Oktober 2007 16:28 An: ids@iiug.org Betreff: Re: AW: chunks status is offline [10140] Hi Marcus, Thanks for your valuable reply. Here is the output of below command: bash-2.03$ onspaces -s dbs01 -p /informix/ol_insrvdb05/dbs01m -o 0 -OWARNING: bringing a down chunk into recovery state. Do you really want to continue? (y/n)y Verifying physical disk space, please wait ... Error opening file /informix/ol_insrvdb05/dbs01m. bash-2.03$ onspaces -s dbs02 -p /informix/ol_insrvdb05/dbs02 -o 0 -OWARNING: bringing a down chunk into recovery state. Do you really want to continue? (y/n)y Verifying physical disk space, please wait ... The chunk '/informix/ol_insrvdb05/dbs02' will not fit in the space specified. The chunks are created on raw devices: lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs01 -> /dev/md/rdsk/d12 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs01m -> /dev/md/rdsk/d20 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs02 -> /dev/md/rdsk/d21 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs02m -> /dev/md/rdsk/d41 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs03 -> /dev/md/rdsk/d31 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs03m -> /dev/md/rdsk/d61 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs04 -> /dev/md/rdsk/d22 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs04m -> /dev/md/rdsk/d32 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs05 -> /dev/md/rdsk/d52 lrwxrwxrwx 1 root other 16 Dec 13 2005 dbs05m -> /dev/md/rdsk/d42 lrwxrwxrwx 1 root other 16 Dec 13 2005 er_dbs1 -> /dev/md/rdsk/d40 lrwxrwxrwx 1 root other 16 Dec 13 2005 er_dbs1m -> /dev/md/rdsk/d51 lrwxrwxrwx 1 root other 16 Dec 13 2005 er_sbdbs1 -> /dev/md/rdsk/d50 lrwxrwxrwx 1 root other 16 Dec 13 2005 er_sbdbs1m -> /dev/md/rdsk/d60 lrwxrwxrwx 1 root other 16 Dec 13 2005 root -> /dev/md/rdsk/d10 lrwxrwxrwx 1 root other 16 Dec 13 2005 rootm -> /dev/md/rdsk/d30 lrwxrwxrwx 1 root other 16 Dec 13 2005 temp1 -> /dev/md/rdsk/d11 Now kindly suggest how to check the permissions and how to proceed further. ************************************************************************ ******* Forum Note: Use "Reply" to post a response in the discussion forum.

Related threads

Glossary