Sybase IQ instance crashing on InfoScale 7.4.1/rhel8 when accessing tempDB volumes which are Veritas VxVM volumes residing on NVME local disks.

book

Article ID: 100062246

calendar_today

Updated On:

Description

Error Message

I. 11/06 20:43:12. 0000000000 {140656154822400} temp Bufman: Decompression error during Read; from slib/s_buf.cxx:2220 into buffer 0x7ff2b85b45c0
I. 11/06 20:43:12. 0000000000 {140656154822400} expected: btype=10,pbn=83843711,txn=0,sp=0,nlb=16,npb=7,iscmp=1,uid=12347745880,blk=70480,mn=2,upmn=2,flgs=0x80000010
I. 11/06 20:43:12. 0000000000 {140656154822400} header: btype=10,pbn=83843711,txn=0,sp=0,nlb=16,npb=7,iscmp=1,ctr=136,cta=134,ver=1,pb=88,pus=4944
I. 11/06 20:43:12. 0000000000 {140656154822400} buffer: 0x0a001000070088867f5aff040000000000000000000000000000000001585013
I. 11/06 20:43:12. 0000000000 {140656154822400} Expected Pad1: 88 (0x58), Pad2: 4944 (0x1350)
I. 11/06 20:43:12. 0000000000 {140656154822400} Got Pad1: 88 (0x58), Pad2: 4944 (0x1350)
I. 11/06 20:43:12. 0000000000  ** slib/s_buf.cxx:2220
I. 11/06 20:43:12. 0000000000   **  Current command:  NO COMMAND OR CURSOR ACTIVE
I. 11/06 20:43:12. 0000000000
I. 11/06 20:43:12. 0000000000     ********************************************************
I. 11/06 20:43:12. 0000000000     **
I. 11/06 20:43:12. 0000000000     **       Non-Fatal IQ Internal Error Detected
I. 11/06 20:43:12. 0000000000     **       at slib/s_buf.cxx:2220 on thread 140656154822400 (TIX 3590)
I. 11/06 20:43:12. 0000000000     **
I. 11/06 20:43:12. 0000000000     **       Please report this to SAP IQ support
I. 11/06 20:43:12. 0000000000     **       with the following diagnostic information,
I. 11/06 20:43:12. 0000000000     **       and, if possible, with the user's command.
I. 11/06 20:43:12. 0000000000     **
I. 11/06 20:43:12. 0000000000     **  Error from IQ connection:
I. 11/06 20:43:12. 0000000000     **  Time of error:  2023-11-06 20:43:12
I. 11/06 20:43:12. 0000000000     **  IQ Version:  SAP IQ/16.1.050/13176/P/SP05.09
I. 11/06 20:43:12. 0000000000     **  OS info:  IQ built on: Linux/Linux64 - x86_64 - 3.10.0-327,  Executed on: Linux/systemb/4.18.0-372.52.1.el8_6.x86_64/#1 SMP Fri Mar 31 06:22:44 EDT 2023/x86_64
I. 11/06 20:43:12. 0000000000     **  Command status when error occurred:  NO COMMAND OR CURSOR ACTIVE
I. 11/06 20:43:12. 0000000000 ***************** This is the STACKTRACE ***************
I. 11/06 20:43:12. 0000000000 ===== Thread Number 140656154822400 =====
I. 11/06 20:43:12. 0000000000 pc: 0x400e81
I. 11/06 20:43:12. 0000000000 pc: 0x7feef5f6eef4 pcstkwalk(stk_trace*, int, db_log*, hos_fd*)+34
I. 11/06 20:43:12. 0000000000 pc: 0x7feef5f6f620 ucstkgentrace(int, int)+190
I. 11/06 20:43:12. 0000000000 pc: 0x7feef5f70546 StackTraceForThisThread(char const*, int, char const*, char const*, char const*)+2c6
I. 11/06 20:43:12. 0000000000 pc: 0x7feef6102a67 s_buf::DumpOnBufmanError(s_bufman_error, char const*, int, short, unsigned long long, unsigned int, unsigned int, unsigned long long, unsigned int, int, s_diskblockheader const*)+17d7
I. 11/06 20:43:12. 0000000000 pc: 0x7feef6105e57 s_buf::CompletePostReadProcessing(s_bufman_errorInfo*, s_diskblockheader*, unsigned long long, unsigned int, unsigned int, int, short, unsigned long long, unsigned int, int, s_diskblockheader*)+4b7
I. 11/06 20:43:12. 0000000000 pc: 0x7feef610ab0f s_bufAioCompletionCallBack::AioCompleted()+cf
I. 11/06 20:43:12. 0000000000 pc: 0x7feef5f3278c hos_aioSystemRequest::Execute()+1ac
I. 11/06 20:43:12. 0000000000 pc: 0x7feef5f327fc hos_aioCallback::AioCompletion(hos_aio const*)+5c
I. 11/06 20:43:12. 0000000000 pc: 0x7feef8155343 hos_aiomgr::ProcessCompletedIOs()+d3
I. 11/06 20:43:12. 0000000000 pc: 0x7feef81553d0 hos_aiomgr::CompletionThread()+50
I. 11/06 20:43:12. 0000000000 pc: 0x7feef5f64999 hos_thread::Main()+209
I. 11/06 20:43:12. 0000000000 pc: 0x7feef5ed8319 hos_lwtask::Start(hos_lwtask*)+69
I. 11/06 20:43:12. 0000000000 pc: 0x7ffff6ab14fe IQWorkerStarter(void*)+3e
I. 11/06 20:43:12. 0000000000 pc: 0x7ffff71c793e run_task_body+1e
I. 11/06 20:43:12. 0000000000 pc: 0x7ffff6d817b8 UnixTask::pre_body(void*)+78
I. 11/06 20:43:12. 0000000000 pc: 0x7ffff65421cf
I. 11/06 20:43:12. 0000000000 pc: 0x7ffff5893dd3
I. 11/06 20:43:12. 0000000000 ******************* End of STACKTRACE ******************
I. 11/06 20:43:12. 0000000000     **
I. 11/06 20:43:12. 0000000000     **       End of stack trace from Non-Fatal IQ Internal Error slib/s_buf.cxx:2220
I. 11/06 20:43:12. 0000000000     **
I. 11/06 20:43:12. 0000000000     ********************************************************
I. 11/06 20:43:12. 0000000000
I. 11/06 20:43:12. 0000000000 INFO [hos_InMemMsgLog::BroadcastDumpInMemMsgLog] from slib/s_buf.cxx:3870 local/remote server ids:1/2
I. 11/06 20:43:12. 0000000000 INFO [hos_InMemMsgLog::BroadcastDumpInMemMsgLog] from slib/s_buf.cxx:3870 local/remote server ids:1/3
I. 11/06 20:43:12. 0000000000 140656154822400 [MpxTrace-INC]-inclib/inc_rpcCmdHandler.cxx:00324 inc_rpcCmdHandler::Connect public & private interconnect is not alive for server 3
I. 11/06 20:43:12. 0000000000 INFO [hos_InMemMsgLog::BroadcastDumpInMemMsgLog] from slib/s_buf.cxx:3870 local/remote server ids:1/4
I. 11/06 20:43:12. 0000000000 INFO [hos_InMemMsgLog::BroadcastDumpInMemMsgLog] from slib/s_buf.cxx:3870 local/remote server ids:1/5
I. 11/06 20:43:12. 0000000000 Page dump for sbuf @ 0x7ff2b85b45c0
I. 11/06 20:43:12. 0000000000  _dskBlk @ 0x7f1270a64000
I. 11/06 20:43:12. 0000000000  Physical block #: 83843711 (0x4ff5a7f)
I. 11/06 20:43:12. 0000000000  Blocks read: 7
I. 11/06 20:43:12. 0000000000  Expected Pad1: 88 (0x58), Pad2: 4944 (0x 1350)
I. 11/06 20:43:12. 0000000000  File #9 (catid 16449): /home/sybase/devices/lv_T09
I. 11/06 20:43:12. 0000000000  Starting Block: 83676160 (0x4fccc00)
I. 11/06 20:43:12. 0000000000  File Physical Block Number: 167552 (0x28e80)
I. 11/06 20:43:12. 0000000000  File size: 9830396 blocks, 161061208064 bytes.
I. 11/06 20:43:12. 0000000000                  | I I N I I H I I I I I I I I I I H I H I |
I. 11/06 20:43:12. 0000000000          Flags:  | s s e s s a s s n s s s s s s s a n e s |
I. 11/06 20:43:12. 0000000000                  | D D e V B s B C W S P U M Q S F s A a P |
I. 11/06 20:43:12. 0000000000                  | i i d e u B l h a p r s a u t r A f d r |
I. 11/06 20:43:12. 0000000000                  | r r s r s u o e s e e e p a a e f f O e |
I. 11/06 20:43:12. 0000000000                  | t t D s y s c c h c f r   r l e f i f A |
I. 11/06 20:43:12. 0000000000                  | i y e i   y k S A i e L   a e l i n T l |
I. 11/06 20:43:12. 0000000000                  | a   a o   W M u r a t o   n   i l n x i |
I. 11/06 20:43:12. 0000000000                  | b   l n   a a m e l c c   t   s i t n o |
I. 11/06 20:43:12. 0000000000        s_buf(s)  | l   l e   i p V a F h k   i   t t y L c |
I. 11/06 20:43:12. 0000000000        follow... | e   o d   t p a   L e e   n   P y A i a |
I. 11/06 20:43:12. 0000000000                  |     c     e e l   B d d   e   a   r s t |
I. 11/06 20:43:12. 0000000000                  |           r d i   u       d   g   e t e |
I. 11/06 20:43:12. 0000000000                  |           s   d   f           e   a   d |
I. 11/06 20:43:12. 0000000000 +----------------+-----------------------------------------+------------+------------+------------+------------+------------+------------+--------------+----------------+----------------+
I. 11/06 20:43:12. 0000000000 |  &s_buf        | A D N V Y W B C I S P U M Q X F Z R H L | LockRefCnt | DBSpaceID  | BlkMapUID  | Blk Num    | _dskBlk    | BlockMap   | VersionRange | prev           | next           |
I. 11/06 20:43:12. 0000000000 +----------------+-----------------------------------------+------------+------------+------------+------------+------------+------------+--------------+----------------+----------------+
I. 11/06 20:43:12. 0000000000   0x7ff2b85b45c0 | a d n v Y w B c i s p U m q x f z r h L |          1 |          0 | 12347745880 |      70480 | 0x7f1270a64000 | 0000000000 |   2..2 |                0 |                0
I. 11/06 20:43:12. 0000000000 [0] Lock{slib/s_bufman.cxx:6371} UnLock{slib/s_blockmap.cxx:6241}
I. 11/06 20:43:12. 0000000000 ---------------------+----------------------+----------------------+------------+-----+----------------------+------------+-------+-------+----------------------
I. 11/06 20:43:12. 0000000000 &s_dskbkhdr          | Block Type           | Transaction ID       | SavePoint  | Ver | Physical Block       | Data       |NumPBlk|NumLBlk| Req/Actual Comp Types
I. 11/06 20:43:12. 0000000000 ---------------------+----------------------+----------------------+------------+-----+----------------------+------------+-------+-------+----------------------
I. 11/06 20:43:12. 0000000000       0x7f1270a64000 |                 sort |                    0 |          0 |   1 |             83843711 | 0x7f1270a64020 |     7 |    16 | ANY/LZRW1
I. 11/06 20:43:12. 0000000000  PBN=83843711 (0x4ff5a7f)

 

 

Cause

As part of changes made to detect NVME devices through IOCTL, an extra buflen was sent to nvme ioctl through the VRTSaslapm component. This bug (e4118605 ) can lead to memory corruption and may cause a system crash.
 

Resolution

A fix is now available for this issue in the current version(s) of the product(s) mentioned. Refer to the link below to obtain the VRTSaslapm package needed to resolve the issue.

The code to resolve this memory corruption is present in VRTSaslapm 7.4.1.3600 which is available on the Veritas Download Center.

 

Issue/Introduction

Sybase IQ for instance, crashes on InfoScale 7.4.1/rhel8 when accessing tempDB volumes, which are Veritas VxVM volumes residing on NVME local disks. When the VxVM volumes were on non-NVME disks or when Linux rawdevices were created on NVME disks, the crashes didn't occur.

Additional Information

JIRA: STESC-8438 ETrack: 4118605