From a602bea3e7ccc5ce3da61d2c18245c4058983926 Mon Sep 17 00:00:00 2001 From: Bryan Schumaker Date: Thu, 26 Jan 2012 16:54:25 -0500 Subject: NFS: Update idmapper documentation Signed-off-by: Bryan Schumaker Signed-off-by: Trond Myklebust --- Documentation/filesystems/nfs/idmapper.txt | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) (limited to 'Documentation') diff --git a/Documentation/filesystems/nfs/idmapper.txt b/Documentation/filesystems/nfs/idmapper.txt index 120fd3cf7fd..fe03d10bb79 100644 --- a/Documentation/filesystems/nfs/idmapper.txt +++ b/Documentation/filesystems/nfs/idmapper.txt @@ -4,13 +4,21 @@ ID Mapper ========= Id mapper is used by NFS to translate user and group ids into names, and to translate user and group names into ids. Part of this translation involves -performing an upcall to userspace to request the information. Id mapper will -user request-key to perform this upcall and cache the result. The program -/usr/sbin/nfs.idmap should be called by request-key, and will perform the -translation and initialize a key with the resulting information. +performing an upcall to userspace to request the information. There are two +ways NFS could obtain this information: placing a call to /sbin/request-key +or by placing a call to the rpc.idmap daemon. + +NFS will attempt to call /sbin/request-key first. If this succeeds, the +result will be cached using the generic request-key cache. This call should +only fail if /etc/request-key.conf is not configured for the id_resolver key +type, see the "Configuring" section below if you wish to use the request-key +method. + +If the call to /sbin/request-key fails (if /etc/request-key.conf is not +configured with the id_resolver key type), then the idmapper will ask the +legacy rpc.idmap daemon for the id mapping. This result will be stored +in a custom NFS idmap cache. - NFS_USE_NEW_IDMAPPER must be selected when configuring the kernel to use this - feature. =========== Configuring -- cgit v1.2.3-70-g09d2 From ef159e9177cc5a09e6174796dde0b2d243ddf28b Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 6 Feb 2012 19:50:40 -0500 Subject: NFSv4.1: Add a module parameter to set the number of session slots Add the module parameter 'max_session_slots' to set the initial number of slots that the NFSv4.1 client will attempt to negotiate with the server. Signed-off-by: Trond Myklebust --- Documentation/kernel-parameters.txt | 8 ++++++++ fs/nfs/nfs4proc.c | 8 +++++++- include/linux/nfs_fs_sb.h | 5 +++-- 3 files changed, 18 insertions(+), 3 deletions(-) (limited to 'Documentation') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 033d4e69b43..1d369c6286e 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1657,6 +1657,14 @@ bytes respectively. Such letter suffixes can also be entirely omitted. of returning the full 64-bit number. The default is to return 64-bit inode numbers. + nfs.max_session_slots= + [NFSv4.1] Sets the maximum number of session slots + the client will attempt to negotiate with the server. + This limits the number of simultaneous RPC requests + that the client can send to the NFSv4.1 server. + Note that there is little point in setting this + value higher than the max_tcp_slot_table_limit. + nfs.nfs4_disable_idmapping= [NFSv4] When set to the default of '1', this option ensures that both the RPC level authentication diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index f3f56f4a3b7..0b331654173 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -72,6 +72,8 @@ #define NFS4_MAX_LOOP_ON_RECOVER (10) +static unsigned short max_session_slots = NFS4_DEF_SLOT_TABLE_SIZE; + struct nfs4_opendata; static int _nfs4_proc_open(struct nfs4_opendata *data); static int _nfs4_recover_proc_open(struct nfs4_opendata *data); @@ -5245,7 +5247,7 @@ static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args) args->fc_attrs.max_rqst_sz = mxrqst_sz; args->fc_attrs.max_resp_sz = mxresp_sz; args->fc_attrs.max_ops = NFS4_MAX_OPS; - args->fc_attrs.max_reqs = session->clp->cl_rpcclient->cl_xprt->max_reqs; + args->fc_attrs.max_reqs = max_session_slots; dprintk("%s: Fore Channel : max_rqst_sz=%u max_resp_sz=%u " "max_ops=%u max_reqs=%u\n", @@ -6390,6 +6392,10 @@ const struct xattr_handler *nfs4_xattr_handlers[] = { NULL }; +module_param(max_session_slots, ushort, 0644); +MODULE_PARM_DESC(max_session_slots, "Maximum number of outstanding NFSv4.1 " + "requests the client will negotiate"); + /* * Local variables: * c-basic-offset: 8 diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index 2ae57f2f645..3bf47666646 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -190,13 +190,14 @@ struct nfs_server { /* maximum number of slots to use */ -#define NFS4_MAX_SLOT_TABLE (128U) +#define NFS4_DEF_SLOT_TABLE_SIZE (16U) +#define NFS4_MAX_SLOT_TABLE (256U) #define NFS4_NO_SLOT ((u32)-1) #if defined(CONFIG_NFS_V4) /* Sessions */ -#define SLOT_TABLE_SZ (NFS4_MAX_SLOT_TABLE/(8*sizeof(long))) +#define SLOT_TABLE_SZ DIV_ROUND_UP(NFS4_MAX_SLOT_TABLE, 8*sizeof(long)) struct nfs4_slot_table { struct nfs4_slot *slots; /* seqid per slot */ unsigned long used_slots[SLOT_TABLE_SZ]; /* used/unused bitmap */ -- cgit v1.2.3-70-g09d2 From db8ac8ba871ae7b97118cfb2913b4986867f09a7 Mon Sep 17 00:00:00 2001 From: Weston Andros Adamson Date: Fri, 17 Feb 2012 15:20:24 -0500 Subject: NFSv4: Send implementation id with exchange_id Send the nfs implementation id in EXCHANGE_ID requests unless the module parameter nfs.send_implementation_id is 0. This adds a CONFIG variable for the nii_domain that defaults to "kernel.org". Signed-off-by: Weston Andros Adamson Signed-off-by: Trond Myklebust --- Documentation/kernel-parameters.txt | 9 ++++++++ fs/nfs/Kconfig | 12 +++++++++++ fs/nfs/nfs4xdr.c | 41 +++++++++++++++++++++++++++++++++++-- 3 files changed, 60 insertions(+), 2 deletions(-) (limited to 'Documentation') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 1d369c6286e..7bae0fd3b63 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1678,6 +1678,15 @@ bytes respectively. Such letter suffixes can also be entirely omitted. back to using the idmapper. To turn off this behaviour, set the value to '0'. + nfs.send_implementation_id = + [NFSv4.1] Send client implementation identification + information in exchange_id requests. + If zero, no implementation identification information + will be sent. + The default is to send the implementation identification + information. + + nmi_debug= [KNL,AVR32,SH] Specify one or more actions to take when a NMI is triggered. Format: [state][,regs][,debounce][,die] diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig index ee86cfcd6c3..7bce64c7060 100644 --- a/fs/nfs/Kconfig +++ b/fs/nfs/Kconfig @@ -99,6 +99,18 @@ config PNFS_OBJLAYOUT depends on NFS_FS && NFS_V4_1 && SCSI_OSD_ULD default m +config NFS_V4_1_IMPLEMENTATION_ID_DOMAIN + string "NFSv4.1 Implementation ID Domain" + depends on NFS_V4_1 + default "kernel.org" + help + This option defines the domain portion of the implementation ID that + may be sent in the NFS exchange_id operation. The value must be in + the format of a DNS domain name and should be set to the DNS domain + name of the distribution. + If the NFS client is unchanged from the upstream kernel, this + option should be set to the default "kernel.org". + config ROOT_NFS bool "Root file system on NFS" depends on NFS_FS=y && IP_PNP diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index ae783436671..d824aedb123 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -44,6 +44,8 @@ #include #include #include +#include +#include #include #include #include @@ -271,7 +273,12 @@ static int nfs4_stat_to_errno(int); 1 /* flags */ + \ 1 /* spa_how */ + \ 0 /* SP4_NONE (for now) */ + \ - 1 /* zero implemetation id array */) + 1 /* implementation id array of size 1 */ + \ + 1 /* nii_domain */ + \ + XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + \ + 1 /* nii_name */ + \ + XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + \ + 3 /* nii_date */) #define decode_exchange_id_maxsz (op_decode_hdr_maxsz + \ 2 /* eir_clientid */ + \ 1 /* eir_sequenceid */ + \ @@ -838,6 +845,12 @@ const u32 nfs41_maxread_overhead = ((RPC_MAX_HEADER_WITH_AUTH + XDR_UNIT); #endif /* CONFIG_NFS_V4_1 */ +static unsigned short send_implementation_id = 1; + +module_param(send_implementation_id, ushort, 0644); +MODULE_PARM_DESC(send_implementation_id, + "Send implementation ID with NFSv4.1 exchange_id"); + static const umode_t nfs_type2fmt[] = { [NF4BAD] = 0, [NF4REG] = S_IFREG, @@ -1766,6 +1779,8 @@ static void encode_exchange_id(struct xdr_stream *xdr, struct compound_hdr *hdr) { __be32 *p; + char impl_name[NFS4_OPAQUE_LIMIT]; + int len = 0; p = reserve_space(xdr, 4 + sizeof(args->verifier->data)); *p++ = cpu_to_be32(OP_EXCHANGE_ID); @@ -1776,7 +1791,29 @@ static void encode_exchange_id(struct xdr_stream *xdr, p = reserve_space(xdr, 12); *p++ = cpu_to_be32(args->flags); *p++ = cpu_to_be32(0); /* zero length state_protect4_a */ - *p = cpu_to_be32(0); /* zero length implementation id array */ + + if (send_implementation_id && + sizeof(CONFIG_NFS_V4_1_IMPLEMENTATION_ID_DOMAIN) > 1 && + sizeof(CONFIG_NFS_V4_1_IMPLEMENTATION_ID_DOMAIN) + <= NFS4_OPAQUE_LIMIT + 1) + len = snprintf(impl_name, sizeof(impl_name), "%s %s %s %s", + utsname()->sysname, utsname()->release, + utsname()->version, utsname()->machine); + + if (len > 0) { + *p = cpu_to_be32(1); /* implementation id array length=1 */ + + encode_string(xdr, + sizeof(CONFIG_NFS_V4_1_IMPLEMENTATION_ID_DOMAIN) - 1, + CONFIG_NFS_V4_1_IMPLEMENTATION_ID_DOMAIN); + encode_string(xdr, len, impl_name); + /* just send zeros for nii_date - the date is in nii_name */ + p = reserve_space(xdr, 12); + p = xdr_encode_hyper(p, 0); + *p = cpu_to_be32(0); + } else + *p = cpu_to_be32(0); /* implementation id array length=0 */ + hdr->nops++; hdr->replen += decode_exchange_id_maxsz; } -- cgit v1.2.3-70-g09d2 From 18d98f6c04991dd3c12acf6f39cea40e9510640a Mon Sep 17 00:00:00 2001 From: Sachin Bhamare Date: Mon, 19 Mar 2012 20:47:58 -0700 Subject: pnfs-obj: autologin: Add support for protocol autologin The pnfs-objects protocol mandates that we autologin into devices not present in the system, according to information specified in the get_device_info returned from the server. The Protocol specifies two login hints. 1. An IP address:port combination 2. A string URI which is constructed as a URL with a protocol prefix followed by :// and a string as address. For each protocol prefix the string-address format might be different. We only support the second option. The first option is just redundant to the second one. NOTE: The Kernel part of autologin does not parse the URI string. It just channels it to a user-mode script. So any new login protocols should only update the user-mode script which is a part of the nfs-utils package, but the Kernel need not change. We implement the autologin by using the call_usermodehelper() API. (Thanks to Steve Dickson for pointing it out) So there is no running daemon needed, and/or special setup. We Add the osd_login_prog Kernel module parameters which defaults to: /sbin/osd_login Kernel try's to upcall the program specified in osd_login_prog. If the file is not found or the execution fails Kernel will disable any farther upcalls, by zeroing out osd_login_prog, Until Admin re-enables it by setting the osd_login_prog parameter to a proper program. Also add text about the osd_login program command line API to: Documentation/filesystems/nfs/pnfs.txt and documentation of the new osd_login_prog module parameter to: Documentation/kernel-parameters.txt TODO: Add timeout option in the case osd_login program gets stuck Signed-off-by: Sachin Bhamare Signed-off-by: Boaz Harrosh Signed-off-by: Trond Myklebust --- Documentation/filesystems/nfs/pnfs.txt | 54 +++++++++++++ Documentation/kernel-parameters.txt | 6 ++ fs/nfs/objlayout/objio_osd.c | 9 +++ fs/nfs/objlayout/objlayout.c | 134 +++++++++++++++++++++++++++++++++ fs/nfs/objlayout/objlayout.h | 2 + 5 files changed, 205 insertions(+) (limited to 'Documentation') diff --git a/Documentation/filesystems/nfs/pnfs.txt b/Documentation/filesystems/nfs/pnfs.txt index 983e14abe7e..c7919c6e3be 100644 --- a/Documentation/filesystems/nfs/pnfs.txt +++ b/Documentation/filesystems/nfs/pnfs.txt @@ -53,3 +53,57 @@ lseg maintains an extra reference corresponding to the NFS_LSEG_VALID bit which holds it in the pnfs_layout_hdr's list. When the final lseg is removed from the pnfs_layout_hdr's list, the NFS_LAYOUT_DESTROYED bit is set, preventing any new lsegs from being added. + +layout drivers +-------------- + +PNFS utilizes what is called layout drivers. The STD defines 3 basic +layout types: "files" "objects" and "blocks". For each of these types +there is a layout-driver with a common function-vectors table which +are called by the nfs-client pnfs-core to implement the different layout +types. + +Files-layout-driver code is in: fs/nfs/nfs4filelayout.c && nfs4filelayoutdev.c +Objects-layout-deriver code is in: fs/nfs/objlayout/.. directory +Blocks-layout-deriver code is in: fs/nfs/blocklayout/.. directory + +objects-layout setup +-------------------- + +As part of the full STD implementation the objlayoutdriver.ko needs, at times, +to automatically login to yet undiscovered iscsi/osd devices. For this the +driver makes up-calles to a user-mode script called *osd_login* + +The path_name of the script to use is by default: + /sbin/osd_login. +This name can be overridden by the Kernel module parameter: + objlayoutdriver.osd_login_prog + +If Kernel does not find the osd_login_prog path it will zero it out +and will not attempt farther logins. An admin can then write new value +to the objlayoutdriver.osd_login_prog Kernel parameter to re-enable it. + +The /sbin/osd_login is part of the nfs-utils package, and should usually +be installed on distributions that support this Kernel version. + +The API to the login script is as follows: + Usage: $0 -u -o -s + Options: + -u target uri e.g. iscsi://: + (allways exists) + (More protocols can be defined in the future. + The client does not interpret this string it is + passed unchanged as recieved from the Server) + -o osdname of the requested target OSD + (Might be empty) + (A string which denotes the OSD name, there is a + limit of 64 chars on this string) + -s systemid of the requested target OSD + (Might be empty) + (This string, if not empty is always an hex + representation of the 20 bytes osd_system_id) + +blocks-layout setup +------------------- + +TODO: Document the setup needs of the blocks layout driver diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 7bae0fd3b63..0d79a88f4de 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1687,6 +1687,12 @@ bytes respectively. Such letter suffixes can also be entirely omitted. information. + objlayoutdriver.osd_login_prog= + [NFS] [OBJLAYOUT] sets the pathname to the program which + is used to automatically discover and login into new + osd-targets. Please see: + Documentation/filesystems/pnfs.txt for more explanations + nmi_debug= [KNL,AVR32,SH] Specify one or more actions to take when a NMI is triggered. Format: [state][,regs][,debounce][,die] diff --git a/fs/nfs/objlayout/objio_osd.c b/fs/nfs/objlayout/objio_osd.c index 3a621a2fd32..4bff4a3dab4 100644 --- a/fs/nfs/objlayout/objio_osd.c +++ b/fs/nfs/objlayout/objio_osd.c @@ -137,6 +137,7 @@ static int objio_devices_lookup(struct pnfs_layout_hdr *pnfslay, struct objio_dev_ent *ode; struct osd_dev *od; struct osd_dev_info odi; + bool retry_flag = true; int err; ode = _dev_list_find(NFS_SERVER(pnfslay->plh_inode), d_id); @@ -171,10 +172,18 @@ static int objio_devices_lookup(struct pnfs_layout_hdr *pnfslay, goto out; } +retry_lookup: od = osduld_info_lookup(&odi); if (unlikely(IS_ERR(od))) { err = PTR_ERR(od); dprintk("%s: osduld_info_lookup => %d\n", __func__, err); + if (err == -ENODEV && retry_flag) { + err = objlayout_autologin(deviceaddr); + if (likely(!err)) { + retry_flag = false; + goto retry_lookup; + } + } goto out; } diff --git a/fs/nfs/objlayout/objlayout.c b/fs/nfs/objlayout/objlayout.c index 157c47e277e..8d45f1c318c 100644 --- a/fs/nfs/objlayout/objlayout.c +++ b/fs/nfs/objlayout/objlayout.c @@ -37,6 +37,9 @@ * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ +#include +#include +#include #include #include "objlayout.h" @@ -651,3 +654,134 @@ void objlayout_put_deviceinfo(struct pnfs_osd_deviceaddr *deviceaddr) __free_page(odi->page); kfree(odi); } + +enum { + OBJLAYOUT_MAX_URI_LEN = 256, OBJLAYOUT_MAX_OSDNAME_LEN = 64, + OBJLAYOUT_MAX_SYSID_HEX_LEN = OSD_SYSTEMID_LEN * 2 + 1, + OSD_LOGIN_UPCALL_PATHLEN = 256 +}; + +static char osd_login_prog[OSD_LOGIN_UPCALL_PATHLEN] = "/sbin/osd_login"; + +module_param_string(osd_login_prog, osd_login_prog, sizeof(osd_login_prog), + 0600); +MODULE_PARM_DESC(osd_login_prog, "Path to the osd_login upcall program"); + +struct __auto_login { + char uri[OBJLAYOUT_MAX_URI_LEN]; + char osdname[OBJLAYOUT_MAX_OSDNAME_LEN]; + char systemid_hex[OBJLAYOUT_MAX_SYSID_HEX_LEN]; +}; + +static int __objlayout_upcall(struct __auto_login *login) +{ + static char *envp[] = { "HOME=/", + "TERM=linux", + "PATH=/sbin:/usr/sbin:/bin:/usr/bin", + NULL + }; + char *argv[8]; + int ret; + + if (unlikely(!osd_login_prog[0])) { + dprintk("%s: osd_login_prog is disabled\n", __func__); + return -EACCES; + } + + dprintk("%s uri: %s\n", __func__, login->uri); + dprintk("%s osdname %s\n", __func__, login->osdname); + dprintk("%s systemid_hex %s\n", __func__, login->systemid_hex); + + argv[0] = (char *)osd_login_prog; + argv[1] = "-u"; + argv[2] = login->uri; + argv[3] = "-o"; + argv[4] = login->osdname; + argv[5] = "-s"; + argv[6] = login->systemid_hex; + argv[7] = NULL; + + ret = call_usermodehelper(argv[0], argv, envp, UMH_WAIT_PROC); + /* + * Disable the upcall mechanism if we're getting an ENOENT or + * EACCES error. The admin can re-enable it on the fly by using + * sysfs to set the objlayoutdriver.osd_login_prog module parameter once + * the problem has been fixed. + */ + if (ret == -ENOENT || ret == -EACCES) { + printk(KERN_ERR "PNFS-OBJ: %s was not found please set " + "objlayoutdriver.osd_login_prog kernel parameter!\n", + osd_login_prog); + osd_login_prog[0] = '\0'; + } + dprintk("%s %s return value: %d\n", __func__, osd_login_prog, ret); + + return ret; +} + +/* Assume dest is all zeros */ +static void __copy_nfsS_and_zero_terminate(struct nfs4_string s, + char *dest, int max_len, + const char *var_name) +{ + if (!s.len) + return; + + if (s.len >= max_len) { + pr_warn_ratelimited( + "objlayout_autologin: %s: s.len(%d) >= max_len(%d)", + var_name, s.len, max_len); + s.len = max_len - 1; /* space for null terminator */ + } + + memcpy(dest, s.data, s.len); +} + +/* Assume sysid is all zeros */ +static void _sysid_2_hex(struct nfs4_string s, + char sysid[OBJLAYOUT_MAX_SYSID_HEX_LEN]) +{ + int i; + char *cur; + + if (!s.len) + return; + + if (s.len != OSD_SYSTEMID_LEN) { + pr_warn_ratelimited( + "objlayout_autologin: systemid_len(%d) != OSD_SYSTEMID_LEN", + s.len); + if (s.len > OSD_SYSTEMID_LEN) + s.len = OSD_SYSTEMID_LEN; + } + + cur = sysid; + for (i = 0; i < s.len; i++) + cur = hex_byte_pack(cur, s.data[i]); +} + +int objlayout_autologin(struct pnfs_osd_deviceaddr *deviceaddr) +{ + int rc; + struct __auto_login login; + + if (!deviceaddr->oda_targetaddr.ota_netaddr.r_addr.len) + return -ENODEV; + + memset(&login, 0, sizeof(login)); + __copy_nfsS_and_zero_terminate( + deviceaddr->oda_targetaddr.ota_netaddr.r_addr, + login.uri, sizeof(login.uri), "URI"); + + __copy_nfsS_and_zero_terminate( + deviceaddr->oda_osdname, + login.osdname, sizeof(login.osdname), "OSDNAME"); + + _sysid_2_hex(deviceaddr->oda_systemid, login.systemid_hex); + + rc = __objlayout_upcall(&login); + if (rc > 0) /* script returns positive values */ + rc = -ENODEV; + + return rc; +} diff --git a/fs/nfs/objlayout/objlayout.h b/fs/nfs/objlayout/objlayout.h index 8ec34727ed2..880ba086be9 100644 --- a/fs/nfs/objlayout/objlayout.h +++ b/fs/nfs/objlayout/objlayout.h @@ -184,4 +184,6 @@ extern void objlayout_encode_layoutreturn( struct xdr_stream *, const struct nfs4_layoutreturn_args *); +extern int objlayout_autologin(struct pnfs_osd_deviceaddr *deviceaddr); + #endif /* _OBJLAYOUT_H */ -- cgit v1.2.3-70-g09d2