static wait_queue_head_t debug_ctlwq;
char libcfs_debug_file_path_arr[PATH_MAX] = LIBCFS_DEBUG_FILE_PATH_DEFAULT;
+EXPORT_SYMBOL(libcfs_debug_file_path_arr);
/* We need to pass a pointer here, but elsewhere this must be a const */
static char *libcfs_debug_file_path;
lprocfs_ir_factor_seq_write(struct file *file, const char __user *buffer,
size_t count, loff_t *off);
#endif
+
+/* lprocfs_status.c: dump pages on cksum error */
+int lprocfs_checksum_dump_seq_show(struct seq_file *m, void *data);
+ssize_t
+lprocfs_checksum_dump_seq_write(struct file *file, const char __user *buffer,
+ size_t count, loff_t *off);
+
extern int lprocfs_single_release(struct inode *, struct file *);
extern int lprocfs_seq_release(struct inode *, struct file *);
struct obd_export *cl_mgc_mgsexp;
/* checksumming for data sent over the network */
- unsigned int cl_checksum:1; /* 0 = disabled, 1 = enabled */
+ unsigned int cl_checksum:1, /* 0 = disabled, 1 = enabled */
+ cl_checksum_dump:1; /* same */
/* supported checksum types that are worked out at connect time */
__u32 cl_supp_cksum_types;
/* checksum algorithm to be used */
* (for /proc/status only!!) */
obd_no_ir:1, /* no imperative recovery. */
obd_process_conf:1, /* device is processing mgs config */
- obd_uses_nid_stats:1; /* maintain per-client OBD stats */
+ obd_uses_nid_stats:1, /* maintain per-client OBD stats */
+ obd_checksum_dump:1; /* dump pages upon cksum error */
/* use separate field as it is set in interrupt to don't mess with
* protection of other bits using _bh lock */
}
EXPORT_SYMBOL(lprocfs_ir_factor_seq_write);
+int lprocfs_checksum_dump_seq_show(struct seq_file *m, void *data)
+{
+ struct obd_device *obd = m->private;
+
+ LASSERT(obd != NULL);
+ seq_printf(m, "%d\n", obd->obd_checksum_dump);
+ return 0;
+}
+EXPORT_SYMBOL(lprocfs_checksum_dump_seq_show);
+
+ssize_t
+lprocfs_checksum_dump_seq_write(struct file *file, const char __user *buffer,
+ size_t count, loff_t *off)
+{
+ struct seq_file *m = file->private_data;
+ struct obd_device *obd = m->private;
+ int rc;
+ __s64 val;
+
+ LASSERT(obd != NULL);
+ rc = lprocfs_str_to_s64(buffer, count, &val);
+ if (rc)
+ return rc;
+
+ obd->obd_checksum_dump = !!val;
+ return count;
+}
+EXPORT_SYMBOL(lprocfs_checksum_dump_seq_write);
+
int lprocfs_recovery_time_soft_seq_show(struct seq_file *m, void *data)
{
struct obd_device *obd = m->private;
LPROC_SEQ_FOPS_RO_TYPE(ofd, num_exports);
LPROC_SEQ_FOPS_RO_TYPE(ofd, target_instance);
LPROC_SEQ_FOPS_RW_TYPE(ofd, ir_factor);
+LPROC_SEQ_FOPS_RW_TYPE(ofd, checksum_dump);
LPROC_SEQ_FOPS_RW_TYPE(ofd, job_interval);
struct lprocfs_vars lprocfs_ofd_obd_vars[] = {
.fops = &ofd_target_instance_fops },
{ .name = "ir_factor",
.fops = &ofd_ir_factor_fops },
+ { .name = "checksum_dump",
+ .fops = &ofd_checksum_dump_fops },
{ .name = "grant_compat_disable",
.fops = &ofd_grant_compat_disable_fops },
{ .name = "client_cache_count",
}
LPROC_SEQ_FOPS(osc_resend_count);
+static int osc_checksum_dump_seq_show(struct seq_file *m, void *v)
+{
+ struct obd_device *obd = m->private;
+
+ if (obd == NULL)
+ return 0;
+
+ seq_printf(m, "%d\n", obd->u.cli.cl_checksum_dump ? 1 : 0);
+ return 0;
+}
+
+static ssize_t osc_checksum_dump_seq_write(struct file *file,
+ const char __user *buffer,
+ size_t count, loff_t *off)
+{
+ struct obd_device *obd;
+ int rc;
+ __s64 val;
+
+ obd = ((struct seq_file *)file->private_data)->private;
+ if (obd == NULL)
+ return 0;
+
+ rc = lprocfs_str_to_s64(buffer, count, &val);
+ if (rc)
+ return rc;
+
+ obd->u.cli.cl_checksum_dump = (val ? 1 : 0);
+
+ return count;
+}
+LPROC_SEQ_FOPS(osc_checksum_dump);
+
static int osc_contention_seconds_seq_show(struct seq_file *m, void *v)
{
struct obd_device *obd = m->private;
.fops = &osc_checksum_fops },
{ .name = "checksum_type",
.fops = &osc_checksum_type_fops },
+ { .name = "checksum_dump",
+ .fops = &osc_checksum_dump_fops },
{ .name = "resend_count",
.fops = &osc_resend_count_fops },
{ .name = "timeouts",
body->oa.o_flags |= cksum_type_pack(cli->cl_cksum_type);
body->oa.o_valid |= OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
}
- }
+
+ /* Client cksum has been already copied to wire obdo in previous
+ * lustre_set_wire_obdo(), and in the case a bulk-read is being
+ * resent due to cksum error, this will allow Server to
+ * check+dump pages on its side */
+ }
ptlrpc_request_set_replen(req);
CLASSERT(sizeof(*aa) <= sizeof(req->rq_async_args));
RETURN(rc);
}
+char dbgcksum_file_name[PATH_MAX];
+
+static void dump_all_bulk_pages(struct obdo *oa, __u32 page_count,
+ struct brw_page **pga, __u32 server_cksum,
+ __u32 client_cksum)
+{
+ struct file *filp;
+ int rc, i;
+ unsigned int len;
+ char *buf;
+ mm_segment_t oldfs;
+
+ /* will only keep dump of pages on first error for the same range in
+ * file/fid, not during the resends/retries. */
+ snprintf(dbgcksum_file_name, sizeof(dbgcksum_file_name),
+ "%s-checksum_dump-osc-"DFID":[%llu-%llu]-%x-%x",
+ (strncmp(libcfs_debug_file_path_arr, "NONE", 4) != 0 ?
+ libcfs_debug_file_path_arr :
+ LIBCFS_DEBUG_FILE_PATH_DEFAULT),
+ oa->o_valid & OBD_MD_FLFID ? oa->o_parent_seq : 0ULL,
+ oa->o_valid & OBD_MD_FLFID ? oa->o_parent_oid : 0,
+ oa->o_valid & OBD_MD_FLFID ? oa->o_parent_ver : 0,
+ pga[0]->off,
+ pga[page_count-1]->off + pga[page_count-1]->count - 1,
+ client_cksum, server_cksum);
+ filp = filp_open(dbgcksum_file_name,
+ O_CREAT | O_EXCL | O_WRONLY | O_LARGEFILE, 0600);
+ if (IS_ERR(filp)) {
+ rc = PTR_ERR(filp);
+ if (rc == -EEXIST)
+ CDEBUG(D_INFO, "%s: can't open to dump pages with "
+ "checksum error: rc = %d\n", dbgcksum_file_name,
+ rc);
+ else
+ CERROR("%s: can't open to dump pages with checksum "
+ "error: rc = %d\n", dbgcksum_file_name, rc);
+ return;
+ }
+
+ oldfs = get_fs();
+ set_fs(KERNEL_DS);
+ for (i = 0; i < page_count; i++) {
+ len = pga[i]->count;
+ buf = kmap(pga[i]->pg);
+ while (len != 0) {
+ rc = vfs_write(filp, (__force const char __user *)buf,
+ len, &filp->f_pos);
+ if (rc < 0) {
+ CERROR("%s: wanted to write %u but got %d "
+ "error\n", dbgcksum_file_name, len, rc);
+ break;
+ }
+ len -= rc;
+ buf += rc;
+ CDEBUG(D_INFO, "%s: wrote %d bytes\n",
+ dbgcksum_file_name, rc);
+ }
+ kunmap(pga[i]->pg);
+ }
+ set_fs(oldfs);
+
+ rc = ll_vfs_fsync_range(filp, 0, LLONG_MAX, 1);
+ if (rc)
+ CERROR("%s: sync returns %d\n", dbgcksum_file_name, rc);
+ filp_close(filp, NULL);
+ return;
+}
+
static int
-check_write_checksum(struct obdo *oa, const struct lnet_process_id *peer,
- __u32 client_cksum, __u32 server_cksum, int nob,
- size_t page_count, struct brw_page **pga,
- cksum_type_t client_cksum_type)
+check_write_checksum(struct obdo *oa, const lnet_process_id_t *peer,
+ __u32 client_cksum, __u32 server_cksum,
+ struct osc_brw_async_args *aa)
{
__u32 new_cksum;
char *msg;
return 0;
}
- cksum_type = cksum_type_unpack(oa->o_valid & OBD_MD_FLFLAGS ?
- oa->o_flags : 0);
- new_cksum = osc_checksum_bulk(nob, page_count, pga, OST_WRITE,
- cksum_type);
+ if (aa->aa_cli->cl_checksum_dump)
+ dump_all_bulk_pages(oa, aa->aa_page_count, aa->aa_ppga,
+ server_cksum, client_cksum);
+
+ cksum_type = cksum_type_unpack(oa->o_valid & OBD_MD_FLFLAGS ?
+ oa->o_flags : 0);
+ new_cksum = osc_checksum_bulk(aa->aa_requested_nob, aa->aa_page_count,
+ aa->aa_ppga, OST_WRITE, cksum_type);
- if (cksum_type != client_cksum_type)
+ if (cksum_type != cksum_type_unpack(aa->aa_oa->o_flags))
msg = "the server did not use the checksum type specified in "
"the original request - likely a protocol problem";
else if (new_cksum == server_cksum)
msg = "changed in transit AND doesn't match the original - "
"likely false positive due to mmap IO (bug 11742)";
- LCONSOLE_ERROR_MSG(0x132, "BAD WRITE CHECKSUM: %s: from %s inode "DFID
- " object "DOSTID" extent [%llu-%llu]\n",
+ LCONSOLE_ERROR_MSG(0x132, "%s: BAD WRITE CHECKSUM: %s: from %s inode "
+ DFID " object "DOSTID" extent [%llu-%llu], original "
+ "client csum %x (type %x), server csum %x (type %x),"
+ " client csum now %x\n",
+ aa->aa_cli->cl_import->imp_obd->obd_name,
msg, libcfs_nid2str(peer->nid),
oa->o_valid & OBD_MD_FLFID ? oa->o_parent_seq : (__u64)0,
oa->o_valid & OBD_MD_FLFID ? oa->o_parent_oid : 0,
oa->o_valid & OBD_MD_FLFID ? oa->o_parent_ver : 0,
- POSTID(&oa->o_oi), pga[0]->off,
- pga[page_count-1]->off + pga[page_count-1]->count - 1);
- CERROR("original client csum %x (type %x), server csum %x (type %x), "
- "client csum now %x\n", client_cksum, client_cksum_type,
- server_cksum, cksum_type, new_cksum);
+ POSTID(&oa->o_oi), aa->aa_ppga[0]->off,
+ aa->aa_ppga[aa->aa_page_count - 1]->off +
+ aa->aa_ppga[aa->aa_page_count-1]->count - 1,
+ client_cksum, cksum_type_unpack(aa->aa_oa->o_flags),
+ server_cksum, cksum_type, new_cksum);
return 1;
}
if ((aa->aa_oa->o_valid & OBD_MD_FLCKSUM) && client_cksum &&
check_write_checksum(&body->oa, peer, client_cksum,
- body->oa.o_cksum, aa->aa_requested_nob,
- aa->aa_page_count, aa->aa_ppga,
- cksum_type_unpack(aa->aa_oa->o_flags)))
+ body->oa.o_cksum, aa))
RETURN(-EAGAIN);
rc = check_write_rcs(req, aa->aa_requested_nob,aa->aa_nio_count,
}
if (server_cksum != client_cksum) {
+ struct ost_body *clbody;
+ u32 page_count = aa->aa_page_count;
+
+ clbody = req_capsule_client_get(&req->rq_pill,
+ &RMF_OST_BODY);
+ if (cli->cl_checksum_dump)
+ dump_all_bulk_pages(&clbody->oa, page_count,
+ aa->aa_ppga, server_cksum,
+ client_cksum);
+
LCONSOLE_ERROR_MSG(0x133, "%s: BAD READ CHECKSUM: from "
"%s%s%s inode "DFID" object "DOSTID
- " extent [%llu-%llu]\n",
+ " extent [%llu-%llu], client %x, "
+ "server %x, cksum_type %x\n",
req->rq_import->imp_obd->obd_name,
libcfs_nid2str(peer->nid),
via, router,
- body->oa.o_valid & OBD_MD_FLFID ?
- body->oa.o_parent_seq : (__u64)0,
- body->oa.o_valid & OBD_MD_FLFID ?
- body->oa.o_parent_oid : 0,
- body->oa.o_valid & OBD_MD_FLFID ?
- body->oa.o_parent_ver : 0,
+ clbody->oa.o_valid & OBD_MD_FLFID ?
+ clbody->oa.o_parent_seq : 0ULL,
+ clbody->oa.o_valid & OBD_MD_FLFID ?
+ clbody->oa.o_parent_oid : 0,
+ clbody->oa.o_valid & OBD_MD_FLFID ?
+ clbody->oa.o_parent_ver : 0,
POSTID(&body->oa.o_oi),
aa->aa_ppga[0]->off,
- aa->aa_ppga[aa->aa_page_count-1]->off +
- aa->aa_ppga[aa->aa_page_count-1]->count -
- 1);
- CERROR("client %x, server %x, cksum_type %x\n",
- client_cksum, server_cksum, cksum_type);
+ aa->aa_ppga[page_count-1]->off +
+ aa->aa_ppga[page_count-1]->count - 1,
+ client_cksum, server_cksum,
+ cksum_type);
cksum_counter = 0;
aa->aa_oa->o_cksum = client_cksum;
rc = -EAGAIN;
memcpy(ptr2, ptr, len);
memcpy(ptr2, "bad3", min(4, len));
kunmap(np);
+
+ /* LU-8376 to preserve original index for
+ * display in dump_all_bulk_pages() */
+ np->index = BD_GET_KIOV(desc,
+ i).kiov_page->index;
+
BD_GET_KIOV(desc, i).kiov_page = np;
} else {
CERROR("%s: can't alloc page for corruption\n",
memcpy(ptr2, ptr, len);
memcpy(ptr2, "bad4", min(4, len));
kunmap(np);
+
+ /* LU-8376 to preserve original index for
+ * display in dump_all_bulk_pages() */
+ np->index = BD_GET_KIOV(desc,
+ i).kiov_page->index;
+
BD_GET_KIOV(desc, i).kiov_page = np;
} else {
CERROR("%s: can't alloc page for corruption\n",
return cksum;
}
+char dbgcksum_file_name[PATH_MAX];
+
+static void dump_all_bulk_pages(struct obdo *oa, int count,
+ lnet_kiov_t *iov, __u32 server_cksum,
+ __u32 client_cksum)
+{
+ struct file *filp;
+ int rc, i;
+ unsigned int len;
+ char *buf;
+ mm_segment_t oldfs;
+
+ /* will only keep dump of pages on first error for the same range in
+ * file/fid, not during the resends/retries. */
+ snprintf(dbgcksum_file_name, sizeof(dbgcksum_file_name),
+ "%s-checksum_dump-ost-"DFID":[%llu-%llu]-%x-%x",
+ (strncmp(libcfs_debug_file_path_arr, "NONE", 4) != 0 ?
+ libcfs_debug_file_path_arr :
+ LIBCFS_DEBUG_FILE_PATH_DEFAULT),
+ oa->o_valid & OBD_MD_FLFID ? oa->o_parent_seq : (__u64)0,
+ oa->o_valid & OBD_MD_FLFID ? oa->o_parent_oid : 0,
+ oa->o_valid & OBD_MD_FLFID ? oa->o_parent_ver : 0,
+ (__u64)iov[0].kiov_page->index << PAGE_SHIFT,
+ ((__u64)iov[count - 1].kiov_page->index << PAGE_SHIFT) +
+ iov[count - 1].kiov_len - 1, client_cksum, server_cksum);
+ filp = filp_open(dbgcksum_file_name,
+ O_CREAT | O_EXCL | O_WRONLY | O_LARGEFILE, 0600);
+ if (IS_ERR(filp)) {
+ rc = PTR_ERR(filp);
+ if (rc == -EEXIST)
+ CDEBUG(D_INFO, "%s: can't open to dump pages with "
+ "checksum error: rc = %d\n", dbgcksum_file_name,
+ rc);
+ else
+ CERROR("%s: can't open to dump pages with checksum "
+ "error: rc = %d\n", dbgcksum_file_name, rc);
+ return;
+ }
+
+ oldfs = get_fs();
+ set_fs(KERNEL_DS);
+ for (i = 0; i < count; i++) {
+ len = iov[i].kiov_len;
+ buf = kmap(iov[i].kiov_page);
+ while (len != 0) {
+ rc = vfs_write(filp, (__force const char __user *)buf,
+ len, &filp->f_pos);
+ if (rc < 0) {
+ CERROR("%s: wanted to write %u but got %d "
+ "error\n", dbgcksum_file_name, len, rc);
+ break;
+ }
+ len -= rc;
+ buf += rc;
+ CDEBUG(D_INFO, "%s: wrote %d bytes\n",
+ dbgcksum_file_name, rc);
+ }
+ kunmap(iov[i].kiov_page);
+ }
+ set_fs(oldfs);
+
+ rc = ll_vfs_fsync_range(filp, 0, LLONG_MAX, 1);
+ if (rc)
+ CERROR("%s: sync returns %d\n", dbgcksum_file_name, rc);
+ filp_close(filp, NULL);
+ return;
+}
+
+static int check_read_checksum(struct ptlrpc_bulk_desc *desc, struct obdo *oa,
+ const lnet_process_id_t *peer,
+ __u32 client_cksum, __u32 server_cksum,
+ cksum_type_t server_cksum_type)
+{
+ char *msg;
+ cksum_type_t cksum_type;
+
+ /* unlikely to happen and only if resend does not occur due to cksum
+ * control failure on Client */
+ if (unlikely(server_cksum == client_cksum)) {
+ CDEBUG(D_PAGE, "checksum %x confirmed upon retry\n",
+ client_cksum);
+ return 0;
+ }
+
+ if (desc->bd_export->exp_obd->obd_checksum_dump)
+ dump_all_bulk_pages(oa, desc->bd_iov_count,
+ &BD_GET_KIOV(desc, 0), server_cksum,
+ client_cksum);
+
+ cksum_type = cksum_type_unpack(oa->o_valid & OBD_MD_FLFLAGS ?
+ oa->o_flags : 0);
+
+ if (cksum_type != server_cksum_type)
+ msg = "the server may have not used the checksum type specified"
+ " in the original request - likely a protocol problem";
+ else
+ msg = "should have changed on the client or in transit";
+
+ LCONSOLE_ERROR_MSG(0x132, "%s: BAD READ CHECKSUM: %s: from %s inode "
+ DFID " object "DOSTID" extent [%llu-%llu], client returned csum"
+ " %x (type %x), server csum %x (type %x)\n",
+ desc->bd_export->exp_obd->obd_name,
+ msg, libcfs_nid2str(peer->nid),
+ oa->o_valid & OBD_MD_FLFID ? oa->o_parent_seq : 0ULL,
+ oa->o_valid & OBD_MD_FLFID ? oa->o_parent_oid : 0,
+ oa->o_valid & OBD_MD_FLFID ? oa->o_parent_ver : 0,
+ POSTID(&oa->o_oi),
+ (__u64)BD_GET_KIOV(desc, 0).kiov_page->index << PAGE_SHIFT,
+ ((__u64)BD_GET_KIOV(desc,
+ desc->bd_iov_count - 1).kiov_page->index
+ << PAGE_SHIFT) +
+ BD_GET_KIOV(desc, desc->bd_iov_count - 1).kiov_len - 1,
+ client_cksum, cksum_type, server_cksum, server_cksum_type);
+ return 1;
+}
+
int tgt_brw_read(struct tgt_session_info *tsi)
{
struct ptlrpc_request *req = tgt_ses_req(tsi);
cksum_type_t cksum_type =
cksum_type_unpack(body->oa.o_valid & OBD_MD_FLFLAGS ?
body->oa.o_flags : 0);
+
repbody->oa.o_flags = cksum_type_pack(cksum_type);
repbody->oa.o_valid = OBD_MD_FLCKSUM | OBD_MD_FLFLAGS;
repbody->oa.o_cksum = tgt_checksum_bulk(tsi->tsi_tgt, desc,
OST_READ, cksum_type);
CDEBUG(D_PAGE, "checksum at read origin: %x\n",
repbody->oa.o_cksum);
+
+ /* if a resend it could be for a cksum error, so check Server
+ * cksum with returned Client cksum (this should even cover
+ * zero-cksum case) */
+ if ((body->oa.o_valid & OBD_MD_FLFLAGS) &&
+ (body->oa.o_flags & OBD_FL_RECOV_RESEND))
+ check_read_checksum(desc, &body->oa, &req->rq_peer,
+ body->oa.o_cksum,
+ repbody->oa.o_cksum, cksum_type);
} else {
repbody->oa.o_valid = 0;
}
router = libcfs_nid2str(desc->bd_sender);
}
+ if (exp->exp_obd->obd_checksum_dump)
+ dump_all_bulk_pages(&body->oa, desc->bd_iov_count,
+ &BD_GET_KIOV(desc, 0), server_cksum,
+ client_cksum);
+
if (mmap) {
CDEBUG_LIMIT(D_INFO, "client csum %x, server csum %x\n",
client_cksum, server_cksum);
return;
}
- LCONSOLE_ERROR_MSG(0x168, "BAD WRITE CHECKSUM: %s from %s%s%s inode "
+ LCONSOLE_ERROR_MSG(0x168, "%s: BAD WRITE CHECKSUM: from %s%s%s inode "
DFID" object "DOSTID" extent [%llu-%llu"
"]: client csum %x, server csum %x\n",
exp->exp_obd->obd_name, libcfs_id2str(req->rq_peer),
}
run_test 77b "checksum error on client write, read"
+cleanup_77c() {
+ trap 0
+ set_checksums 0
+ $LCTL set_param osc.*osc-[^mM]*.checksum_dump=0
+ $check_ost &&
+ do_facet ost1 $LCTL set_param obdfilter.*-OST*.checksum_dump=0
+ [ -n $osc_file_prefix ] && rm -f ${osc_file_prefix}*
+ $check_ost && [ -n $ost_file_prefix ] &&
+ do_facet ost1 rm -f ${ost_file_prefix}\*
+}
+
+test_77c() {
+ [ $PARALLEL == "yes" ] && skip "skip parallel run" && return
+ $GSS && skip "could not run with gss" && return
+
+ local bad1
+ local osc_file_prefix
+ local osc_file
+ local check_ost=false
+ local ost_file_prefix
+ local ost_file
+ local orig_cksum
+ local dump_cksum
+ local fid
+
+ # ensure corruption will occur on first OSS/OST
+ $LFS setstripe -i 0 $DIR/$tfile
+
+ [ ! -f $F77_TMP ] && setup_f77
+ dd if=$F77_TMP of=$DIR/$tfile bs=1M count=$F77SZ conv=sync ||
+ error "dd write error: $?"
+ fid=$($LFS path2fid $DIR/$tfile)
+
+ if [ $(lustre_version_code ost1) -ge $(version_code 2.5.42.6) ]
+ then
+ check_ost=true
+ ost_file_prefix=$(do_facet ost1 $LCTL get_param -n debug_path)
+ ost_file_prefix=${ost_file_prefix}-checksum_dump-ost-\\${fid}
+ else
+ echo "OSS do not support bulk pages dump upon error"
+ fi
+
+ osc_file_prefix=$($LCTL get_param -n debug_path)
+ osc_file_prefix=${osc_file_prefix}-checksum_dump-osc-\\${fid}
+
+ trap cleanup_77c EXIT
+
+ set_checksums 1
+ # enable bulk pages dump upon error on Client
+ $LCTL set_param osc.*osc-[^mM]*.checksum_dump=1
+ # enable bulk pages dump upon error on OSS
+ $check_ost &&
+ do_facet ost1 $LCTL set_param obdfilter.*-OST*.checksum_dump=1
+
+ # flush Client cache to allow next read to reach OSS
+ cancel_lru_locks osc
+
+ #define OBD_FAIL_OSC_CHECKSUM_RECEIVE 0x408
+ $LCTL set_param fail_loc=0x80000408
+ dd if=$DIR/$tfile of=/dev/null bs=1M || error "dd read error: $?"
+ $LCTL set_param fail_loc=0
+
+ rm -f $DIR/$tfile
+
+ # check cksum dump on Client
+ osc_file=$(ls ${osc_file_prefix}*)
+ [ -n "$osc_file" ] || error "no checksum dump file on Client"
+ # OBD_FAIL_OSC_CHECKSUM_RECEIVE corrupts with "bad1" at start of file
+ bad1=$(dd if=$osc_file bs=1 count=4 2>/dev/null) || error "dd error: $?"
+ [ $bad1 == "bad1" ] || error "unexpected corrupt pattern"
+ orig_cksum=$(dd if=$F77_TMP bs=1 skip=4 count=1048572 2>/dev/null |
+ cksum)
+ dump_cksum=$(dd if=$osc_file bs=1 skip=4 2>/dev/null | cksum)
+ [[ "$orig_cksum" == "$dump_cksum" ]] ||
+ error "dump content does not match on Client"
+
+ $check_ost || skip "No need to check cksum dump on OSS"
+
+ # check cksum dump on OSS
+ ost_file=$(do_facet ost1 ls ${ost_file_prefix}\*)
+ [ -n "$ost_file" ] || error "no checksum dump file on OSS"
+ orig_cksum=$(dd if=$F77_TMP bs=1048576 count=1 2>/dev/null | cksum)
+ dump_cksum=$(do_facet ost1 dd if=$ost_file 2>/dev/null \| cksum)
+ [[ "$orig_cksum" == "$dump_cksum" ]] ||
+ error "dump content does not match on OSS"
+
+ cleanup_77c
+}
+run_test 77c "checksum error on client read with debug"
+
test_77d() { # bug 10889
[ $PARALLEL == "yes" ] && skip "skip parallel run" && return
$GSS && skip "could not run with gss" && return
git://git.whamcloud.com / fs / lustre-release.git / commitdiff