PROBLEM: (QAR 69961) (Patch ID: OSF440-036) ******** This patch corrects a problem where a mount(8) command failure caused the operating system to crash. Instead, the failure will now only cause the AdvFS filesystem domain to shut down. The mount(8) command would fail when mounting an AdvFS file system because of an internal file system error. Now the file system shuts down the AdvFS file domain rather than crashing the system with a panic(). PROBLEM: (XIP100014, XIO100017) (Patch ID: OSF440-056) ******** This patch fixes a problem on systems using the AdvFS filesystem, where the system can panic with the panic string: "del_clean_mcell_list: no primary xtnt record". The stack trace will be similar to: advfs_sad del_clean_mcell_list bs_bfdmn_activate bs_bfdmn_tbl_activate fs_fset_get_info msfs_real_syscall msfs_syscall syscall _Xsyscall PROBLEM: (MGO103845, MGO103895) (Patch ID: OSF440-057) ******** This patch fixes an AdvFS Domain Panic that occurs with the following message on the console: "load_x_cache: bad status from bs_refpg of sbm" Example stack traces are as follows. Note: These are obtained from a system panic and not from the domain panic itself. 5 panic 6 advfs_sad 7 load_x_cache 8 sbm_find_space 9 alloc_from_bitmap 10 alloc_from_one_disk 11 stg_alloc_from_one_disk 12 alloc_copy_stg 13 migrate_normal 14 mig_migrate 15 bs_migrate 16 msfs_real_syscall 17 _Xsyscall PROBLEM: (HPAQC0B99) (Patch ID: OSF440-058) ******** This patch fixes a problem in AdvFS with auditing enabled. During a rmvol operation, the system can panic with a "kernel memory fault" with a stack trace that is similar to the following: > 0 boot() 1 panic(s = 0xfffffc00005b60a8 = "kernel memory fault") 2 trap() 3 _XentMM() 4 audit_rec_build() 5 audit_call() 6 msfs_audit_syscall() 7 msfs_real_syscall() 8 msfs_syscall() 9 syscall() 10 _Xsyscall() PROBLEM: (BCPM117VX, QAR 68804) (Patch ID: OSF440-059) ******** This patch fixes a problem where the "statfs" system call was reporting incorrect block usage on AdvFS filesets. As a side effect of this problem, the sendmail utility may sleep needlessly (waiting for space to become available). PROBLEM: (QAR 58502,QAR 59094,QAR 61399,QAR 67023) (Patch ID: OSF440-060) ******** This patch provides the following fixes and enhancements to AdvFS: - AdvFS volumes were not setting the default I/O byte transfer size to the preferred size reported by the disk drives. This caused I/O consolidations to be smaller than is optimal for the disks. This patch now assigns the driver's preferred I/O transfer byte size to all new AdvFS volumes created with mkfdmn and addvol. - AdvFS chvol read and write transfer size range was increased. With this patch the transfer size range is now between 16 and 3276 512-byte blocks. - The read-ahead algorithm was modified to improve performance under certain conditions. The algorithm now sanity checks the number of pages that it would attempt to prefetch and apply the following restrictions when doing its calculations: o Only read-ahead the number of pages remaining in the file. o Use a predetermined percentage of the buffer pool for use as read-ahead buffers. PROBLEM: (QAR 70283) (Patch ID: OSF440-112) ******** This patch fixes the problem where the system panics if AdvFS detects an inconsistency in the free list of mcells that is kept on a per-volume basis in an AdvFS domain. The panic string seen with this panic is "alloc_mcell: bad mcell free list". PROBLEM: (UVO106494) (Patch ID: OSF440-125) ******** This patch fixes a problem where "update" takes too long to sync mmap files when using an AdvFS file system. PROBLEM: (QARS 70619 68822 72776 72391) (Patch ID: OSF440-141) ******** This patch fixes the following two problems in AdvFS: - When a "log half full" or "log full" problem occurs, an entire system will panic. With this patch, only the AdvFS domain will panic instead of bringing down the entire system. - The erorr message "ftx_bfdmn_recovery:bad record size\n N1 = 1" is received when the wordCnt, as returned by lgr_read, is not enough to hold the ftxDoneLRT record that preceeds each log record in a log page. This caused the log page trailer to overwrite the ftxDoneLRT record in a log record. This patch now leaves enough room to write a log header at the end of a log page. PROBLEM: (CLD ALC-08618, QAR 69288) (Patch ID: OSF440-147) ******** This patch corrects a problem where a "can't clear a bit twice" panic occurs after an unanticipated system crash and an improperly handled AdvFS recovery operation. Note: Although the system will have paniced with a "can't clear a bit twice" panic message, the AdvFS file domain is often still good. On the following reboot the system will not mount the affected AdvFS domain complaining that the domain is in an "inconsistent state". To recover the domain, use dbx to set the variable AdvfsFixUpSBM to 1: dbx dbx>a AdvfsFixUpSBM = 1 dbx>quit Then use the AdvFS verify utility to check the domain. If the recovery is successful, verify will report no errors and the domain will be mountable. Please remember to set AdvfsFixUpSBM back to zero when you have finished recovering the domain. PROBLEM: (MGO104093,MGO104098,DSAT31W43, QAR 71350) (Patch ID: OSF440-082) ******** This patch corrects a problem in AdvFS which causes single-CPU systems to hang and causes multiple-CPU systems to panic with a "simple lock time limit exceeded" message specifying lock class name BfAccessTblMutex. This problem was originally reported when sys_check was being used to report on systems where there are more than 256 AdvFS filesets. The same problem could occur with any form of file access with that many filesets. PROBLEM: (MGO103900, QAR 71550) (Patch ID: OSF440-097) ******** This patch corrects a problem in AdvFS where unmounting a domain that is already in a panicked state could result in the following system panic message: log_flush_sync: pinpg error\n N1 = 5 Note: In this instance an I/O error (EIO = 5) caused the original domain panic. Stack trace: 6 advfs_sad() 7 log_flush_sync() 8 lgr_writev_ftx() 9 log_donerec_nunpin() 10 ftx_done_urdr() 11 ftx_done_n() 12 quota_deactivate() 13 msfs_unmount() Kernel Rebuild required.