/* Copyright (c) 2006-2012 Red Hat, Inc. This file is part of GlusterFS. This file is licensed to you under your choice of the GNU Lesser General Public License, version 3 or any later version (LGPLv3 or later), or the GNU General Public License, version 2 (GPLv2), in all cases as published by the Free Software Foundation. */ #ifndef _CONFIG_H #define _CONFIG_H #include "config.h" #endif #define __XOPEN_SOURCE 500 #include #include #include #include #include #include #include #include #include #ifndef GF_BSD_HOST_OS #include #endif /* GF_BSD_HOST_OS */ #include "glusterfs.h" #include "checksum.h" #include "dict.h" #include "logging.h" #include "posix.h" #include "xlator.h" #include "defaults.h" #include "common-utils.h" #include "compat-errno.h" #include "compat.h" #include "byte-order.h" #include "syscall.h" #include "statedump.h" #include "locking.h" #include "timer.h" #include "glusterfs3-xdr.h" #include "hashfn.h" #include char *marker_xattrs[] = {"trusted.glusterfs.quota.*", "trusted.glusterfs.*.xtime", NULL}; static char* posix_ignore_xattrs[] = { "gfid-req", GLUSTERFS_ENTRYLK_COUNT, GLUSTERFS_INODELK_COUNT, GLUSTERFS_POSIXLK_COUNT, NULL }; gf_boolean_t posix_special_xattr (char **pattern, char *key) { int i = 0; gf_boolean_t flag = _gf_false; GF_VALIDATE_OR_GOTO ("posix", pattern, out); GF_VALIDATE_OR_GOTO ("posix", key, out); for (i = 0; pattern[i]; i++) { if (!fnmatch (pattern[i], key, 0)) { flag = _gf_true; break; } } out: return flag; } static gf_boolean_t posix_xattr_ignorable (char *key, posix_xattr_filler_t *filler) { int i = 0; gf_boolean_t ignore = _gf_false; GF_ASSERT (key); if (!key) goto out; for (i = 0; posix_ignore_xattrs[i]; i++) { if (!strcmp (key, posix_ignore_xattrs[i])) { ignore = _gf_true; goto out; } } if ((!strcmp (key, GF_CONTENT_KEY)) && (!IA_ISREG (filler->stbuf->ia_type))) ignore = _gf_true; out: return ignore; } static int _posix_xattr_get_set (dict_t *xattr_req, char *key, data_t *data, void *xattrargs) { posix_xattr_filler_t *filler = xattrargs; char *value = NULL; ssize_t xattr_size = -1; int ret = -1; char *databuf = NULL; int _fd = -1; loc_t *loc = NULL; ssize_t req_size = 0; if (posix_xattr_ignorable (key, filler)) goto out; /* should size be put into the data_t ? */ if (!strcmp (key, GF_CONTENT_KEY) && IA_ISREG (filler->stbuf->ia_type)) { /* file content request */ req_size = data_to_uint64 (data); if (req_size >= filler->stbuf->ia_size) { _fd = open (filler->real_path, O_RDONLY); if (_fd == -1) { gf_log (filler->this->name, GF_LOG_ERROR, "Opening file %s failed: %s", filler->real_path, strerror (errno)); goto err; } databuf = GF_CALLOC (1, filler->stbuf->ia_size, gf_posix_mt_char); if (!databuf) { goto err; } ret = read (_fd, databuf, filler->stbuf->ia_size); if (ret == -1) { gf_log (filler->this->name, GF_LOG_ERROR, "Read on file %s failed: %s", filler->real_path, strerror (errno)); goto err; } ret = close (_fd); _fd = -1; if (ret == -1) { gf_log (filler->this->name, GF_LOG_ERROR, "Close on file %s failed: %s", filler->real_path, strerror (errno)); goto err; } ret = dict_set_bin (filler->xattr, key, databuf, filler->stbuf->ia_size); if (ret < 0) { gf_log (filler->this->name, GF_LOG_ERROR, "failed to set dict value. key: %s, path: %s", key, filler->real_path); goto err; } /* To avoid double free in cleanup below */ databuf = NULL; err: if (_fd != -1) close (_fd); GF_FREE (databuf); } } else if (!strcmp (key, GLUSTERFS_OPEN_FD_COUNT)) { loc = filler->loc; if (loc) { ret = dict_set_uint32 (filler->xattr, key, loc->inode->fd_count); if (ret < 0) gf_log (filler->this->name, GF_LOG_WARNING, "Failed to set dictionary value for %s", key); } } else { xattr_size = sys_lgetxattr (filler->real_path, key, NULL, 0); if (xattr_size > 0) { value = GF_CALLOC (1, xattr_size + 1, gf_posix_mt_char); if (!value) return -1; xattr_size = sys_lgetxattr (filler->real_path, key, value, xattr_size); if (xattr_size <= 0) { gf_log (filler->this->name, GF_LOG_WARNING, "getxattr failed. path: %s, key: %s", filler->real_path, key); GF_FREE (value); return -1; } value[xattr_size] = '\0'; ret = dict_set_bin (filler->xattr, key, value, xattr_size); if (ret < 0) { gf_log (filler->this->name, GF_LOG_DEBUG, "dict set failed. path: %s, key: %s", filler->real_path, key); GF_FREE (value); } } } out: return 0; } int posix_fill_gfid_path (xlator_t *this, const char *path, struct iatt *iatt) { int ret = 0; ssize_t size = 0; if (!iatt) return 0; size = sys_lgetxattr (path, GFID_XATTR_KEY, iatt->ia_gfid, 16); /* Return value of getxattr */ if ((size == 16) || (size == -1)) ret = 0; else ret = size; return ret; } int posix_fill_gfid_fd (xlator_t *this, int fd, struct iatt *iatt) { int ret = 0; ssize_t size = 0; if (!iatt) return 0; size = sys_fgetxattr (fd, GFID_XATTR_KEY, iatt->ia_gfid, 16); /* Return value of getxattr */ if ((size == 16) || (size == -1)) ret = 0; else ret = size; return ret; } void posix_fill_ino_from_gfid (xlator_t *this, struct iatt *buf) { uint64_t temp_ino = 0; int j = 0; int i = 0; /* consider least significant 8 bytes of value out of gfid */ if (uuid_is_null (buf->ia_gfid)) { buf->ia_ino = -1; goto out; } for (i = 15; i > (15 - 8); i--) { temp_ino += (uint64_t)(buf->ia_gfid[i]) << j; j += 8; } buf->ia_ino = temp_ino; out: return; } int posix_fdstat (xlator_t *this, int fd, struct iatt *stbuf_p) { int ret = 0; struct stat fstatbuf = {0, }; struct iatt stbuf = {0, }; ret = fstat (fd, &fstatbuf); if (ret == -1) goto out; if (fstatbuf.st_nlink && !S_ISDIR (fstatbuf.st_mode)) fstatbuf.st_nlink--; iatt_from_stat (&stbuf, &fstatbuf); ret = posix_fill_gfid_fd (this, fd, &stbuf); if (ret) gf_log_callingfn (this->name, GF_LOG_DEBUG, "failed to get gfid"); posix_fill_ino_from_gfid (this, &stbuf); if (stbuf_p) *stbuf_p = stbuf; out: return ret; } int posix_istat (xlator_t *this, uuid_t gfid, const char *basename, struct iatt *buf_p) { char *real_path = NULL; struct stat lstatbuf = {0, }; struct iatt stbuf = {0, }; int ret = 0; struct posix_private *priv = NULL; priv = this->private; MAKE_HANDLE_PATH (real_path, this, gfid, basename); ret = lstat (real_path, &lstatbuf); if (ret != 0) { if (ret == -1) { if (errno != ENOENT && errno != ELOOP) gf_log (this->name, GF_LOG_WARNING, "lstat failed on %s (%s)", real_path, strerror (errno)); } else { // may be some backend filesystem issue gf_log (this->name, GF_LOG_ERROR, "lstat failed on " "%s and return value is %d instead of -1. " "Please see dmesg output to check whether the " "failure is due to backend filesystem issue", real_path, ret); ret = -1; } goto out; } if ((lstatbuf.st_ino == priv->handledir.st_ino) && (lstatbuf.st_dev == priv->handledir.st_dev)) { errno = ENOENT; return -1; } if (!S_ISDIR (lstatbuf.st_mode)) lstatbuf.st_nlink --; iatt_from_stat (&stbuf, &lstatbuf); if (basename) posix_fill_gfid_path (this, real_path, &stbuf); else uuid_copy (stbuf.ia_gfid, gfid); posix_fill_ino_from_gfid (this, &stbuf); if (buf_p) *buf_p = stbuf; out: return ret; } int posix_pstat (xlator_t *this, uuid_t gfid, const char *path, struct iatt *buf_p) { struct stat lstatbuf = {0, }; struct iatt stbuf = {0, }; int ret = 0; struct posix_private *priv = NULL; priv = this->private; ret = lstat (path, &lstatbuf); if (ret != 0) { if (ret == -1) { if (errno != ENOENT) gf_log (this->name, GF_LOG_WARNING, "lstat failed on %s (%s)", path, strerror (errno)); } else { // may be some backend filesytem issue gf_log (this->name, GF_LOG_ERROR, "lstat failed on " "%s and return value is %d instead of -1. " "Please see dmesg output to check whether the " "failure is due to backend filesystem issue", path, ret); ret = -1; } goto out; } if ((lstatbuf.st_ino == priv->handledir.st_ino) && (lstatbuf.st_dev == priv->handledir.st_dev)) { errno = ENOENT; return -1; } if (!S_ISDIR (lstatbuf.st_mode)) lstatbuf.st_nlink --; iatt_from_stat (&stbuf, &lstatbuf); if (gfid && !uuid_is_null (gfid)) uuid_copy (stbuf.ia_gfid, gfid); else posix_fill_gfid_path (this, path, &stbuf); posix_fill_ino_from_gfid (this, &stbuf); if (buf_p) *buf_p = stbuf; out: return ret; } dict_t * posix_lookup_xattr_fill (xlator_t *this, const char *real_path, loc_t *loc, dict_t *xattr_req, struct iatt *buf) { dict_t *xattr = NULL; posix_xattr_filler_t filler = {0, }; xattr = get_new_dict(); if (!xattr) { goto out; } filler.this = this; filler.real_path = real_path; filler.xattr = xattr; filler.stbuf = buf; filler.loc = loc; dict_foreach (xattr_req, _posix_xattr_get_set, &filler); out: return xattr; } int posix_gfid_set (xlator_t *this, const char *path, loc_t *loc, dict_t *xattr_req) { void *uuid_req = NULL; uuid_t uuid_curr; int ret = 0; ssize_t size = 0; struct stat stat = {0, }; if (!xattr_req) goto out; if (sys_lstat (path, &stat) != 0) goto out; size = sys_lgetxattr (path, GFID_XATTR_KEY, uuid_curr, 16); if (size == 16) { ret = 0; goto verify_handle; } ret = dict_get_ptr (xattr_req, "gfid-req", &uuid_req); if (ret) { gf_log (this->name, GF_LOG_DEBUG, "failed to get the gfid from dict for %s", loc->path); goto out; } ret = sys_lsetxattr (path, GFID_XATTR_KEY, uuid_req, 16, XATTR_CREATE); if (ret != 0) { gf_log (this->name, GF_LOG_WARNING, "setting GFID on %s failed (%s)", path, strerror (errno)); goto out; } uuid_copy (uuid_curr, uuid_req); verify_handle: if (!S_ISDIR (stat.st_mode)) ret = posix_handle_hard (this, path, uuid_curr, &stat); else ret = posix_handle_soft (this, path, loc, uuid_curr, &stat); out: return ret; } int posix_set_file_contents (xlator_t *this, const char *path, char *keyp, data_t *value, int flags) { char * key = NULL; char real_path[PATH_MAX]; int32_t file_fd = -1; int op_ret = 0; int ret = -1; /* XXX: does not handle assigning GFID to created files */ return -1; key = &(keyp[15]); sprintf (real_path, "%s/%s", path, key); if (flags & XATTR_REPLACE) { /* if file exists, replace it * else, error out */ file_fd = open (real_path, O_TRUNC|O_WRONLY); if (file_fd == -1) { goto create; } if (value->len) { ret = write (file_fd, value->data, value->len); if (ret == -1) { op_ret = -errno; gf_log (this->name, GF_LOG_ERROR, "write failed while doing setxattr " "for key %s on path %s: %s", key, real_path, strerror (errno)); goto out; } ret = close (file_fd); if (ret == -1) { op_ret = -errno; gf_log (this->name, GF_LOG_ERROR, "close failed on %s: %s", real_path, strerror (errno)); goto out; } } create: /* we know file doesn't exist, create it */ file_fd = open (real_path, O_CREAT|O_WRONLY, 0644); if (file_fd == -1) { op_ret = -errno; gf_log (this->name, GF_LOG_ERROR, "failed to open file %s with O_CREAT: %s", key, strerror (errno)); goto out; } ret = write (file_fd, value->data, value->len); if (ret == -1) { op_ret = -errno; gf_log (this->name, GF_LOG_ERROR, "write failed on %s while setxattr with " "key %s: %s", real_path, key, strerror (errno)); goto out; } ret = close (file_fd); if (ret == -1) { op_ret = -errno; gf_log (this->name, GF_LOG_ERROR, "close failed on %s while setxattr with " "key %s: %s", real_path, key, strerror (errno)); goto out; } } out: return op_ret; } int posix_get_file_contents (xlator_t *this, uuid_t pargfid, const char *name, char **contents) { char *real_path = NULL; int32_t file_fd = -1; struct iatt stbuf = {0,}; int op_ret = 0; int ret = -1; MAKE_HANDLE_PATH (real_path, this, pargfid, name); op_ret = posix_istat (this, pargfid, name, &stbuf); if (op_ret == -1) { op_ret = -errno; gf_log (this->name, GF_LOG_ERROR, "lstat failed on %s: %s", real_path, strerror (errno)); goto out; } file_fd = open (real_path, O_RDONLY); if (file_fd == -1) { op_ret = -errno; gf_log (this->name, GF_LOG_ERROR, "open failed on %s: %s", real_path, strerror (errno)); goto out; } *contents = GF_CALLOC (stbuf.ia_size + 1, sizeof(char), gf_posix_mt_char); if (! *contents) { op_ret = -errno; goto out; } ret = read (file_fd, *contents, stbuf.ia_size); if (ret <= 0) { op_ret = -1; gf_log (this->name, GF_LOG_ERROR, "read on %s failed: %s", real_path, strerror (errno)); goto out; } *contents[stbuf.ia_size] = '\0'; op_ret = close (file_fd); file_fd = -1; if (op_ret == -1) { op_ret = -errno; gf_log (this->name, GF_LOG_ERROR, "close on %s failed: %s", real_path, strerror (errno)); goto out; } out: if (op_ret < 0) { GF_FREE (*contents); if (file_fd != -1) close (file_fd); } return op_ret; } static int gf_xattr_enotsup_log; int posix_handle_pair (xlator_t *this, const char *real_path, char *key, data_t *value, int flags) { int sys_ret = -1; int ret = 0; if (ZR_FILE_CONTENT_REQUEST(key)) { ret = posix_set_file_contents (this, real_path, key, value, flags); } else { sys_ret = sys_lsetxattr (real_path, key, value->data, value->len, flags); if (sys_ret < 0) { if (errno == ENOTSUP) { GF_LOG_OCCASIONALLY(gf_xattr_enotsup_log, this->name,GF_LOG_WARNING, "Extended attributes not " "supported (try remounting " "brick with 'user_xattr' " "flag)"); } else if (errno == ENOENT) { if (!posix_special_xattr (marker_xattrs, key)) { gf_log (this->name, GF_LOG_ERROR, "setxattr on %s failed: %s", real_path, strerror (errno)); } } else { #ifdef GF_DARWIN_HOST_OS gf_log (this->name, ((errno == EINVAL) ? GF_LOG_DEBUG : GF_LOG_ERROR), "%s: key:%s error:%s", real_path, key, strerror (errno)); #else /* ! DARWIN */ gf_log (this->name, GF_LOG_ERROR, "%s: key:%s error:%s", real_path, key, strerror (errno)); #endif /* DARWIN */ } ret = -errno; goto out; } } out: return ret; } int posix_fhandle_pair (xlator_t *this, int fd, char *key, data_t *value, int flags) { int sys_ret = -1; int ret = 0; sys_ret = sys_fsetxattr (fd, key, value->data, value->len, flags); if (sys_ret < 0) { if (errno == ENOTSUP) { GF_LOG_OCCASIONALLY(gf_xattr_enotsup_log, this->name,GF_LOG_WARNING, "Extended attributes not " "supported (try remounting " "brick with 'user_xattr' " "flag)"); } else if (errno == ENOENT) { gf_log (this->name, GF_LOG_ERROR, "fsetxattr on fd=%d failed: %s", fd, strerror (errno)); } else { #ifdef GF_DARWIN_HOST_OS gf_log (this->name, ((errno == EINVAL) ? GF_LOG_DEBUG : GF_LOG_ERROR), "fd=%d: key:%s error:%s", fd, key, strerror (errno)); #else /* ! DARWIN */ gf_log (this->name, GF_LOG_ERROR, "fd=%d: key:%s error:%s", fd, key, strerror (errno)); #endif /* DARWIN */ } ret = -errno; goto out; } out: return ret; } static int janitor_walker (const char *fpath, const struct stat *sb, int typeflag, struct FTW *ftwbuf) { struct iatt stbuf = {0, }; xlator_t *this = NULL; this = THIS; posix_pstat (this, NULL, fpath, &stbuf); switch (sb->st_mode & S_IFMT) { case S_IFREG: case S_IFBLK: case S_IFLNK: case S_IFCHR: case S_IFIFO: case S_IFSOCK: gf_log (THIS->name, GF_LOG_TRACE, "unlinking %s", fpath); unlink (fpath); if (stbuf.ia_nlink == 1) posix_handle_unset (this, stbuf.ia_gfid, NULL); break; case S_IFDIR: if (ftwbuf->level) { /* don't remove top level dir */ gf_log (THIS->name, GF_LOG_TRACE, "removing directory %s", fpath); rmdir (fpath); posix_handle_unset (this, stbuf.ia_gfid, NULL); } break; } return 0; /* 0 = FTW_CONTINUE */ } static struct posix_fd * janitor_get_next_fd (xlator_t *this) { struct posix_private *priv = NULL; struct posix_fd *pfd = NULL; struct timespec timeout; priv = this->private; pthread_mutex_lock (&priv->janitor_lock); { if (list_empty (&priv->janitor_fds)) { time (&timeout.tv_sec); timeout.tv_sec += priv->janitor_sleep_duration; timeout.tv_nsec = 0; pthread_cond_timedwait (&priv->janitor_cond, &priv->janitor_lock, &timeout); goto unlock; } pfd = list_entry (priv->janitor_fds.next, struct posix_fd, list); list_del (priv->janitor_fds.next); } unlock: pthread_mutex_unlock (&priv->janitor_lock); return pfd; } static void * posix_janitor_thread_proc (void *data) { xlator_t * this = NULL; struct posix_private *priv = NULL; struct posix_fd *pfd; time_t now; this = data; priv = this->private; THIS = this; while (1) { time (&now); if ((now - priv->last_landfill_check) > priv->janitor_sleep_duration) { gf_log (this->name, GF_LOG_TRACE, "janitor cleaning out %s", priv->trash_path); nftw (priv->trash_path, janitor_walker, 32, FTW_DEPTH | FTW_PHYS); priv->last_landfill_check = now; } pfd = janitor_get_next_fd (this); if (pfd) { if (pfd->dir == NULL) { gf_log (this->name, GF_LOG_TRACE, "janitor: closing file fd=%d", pfd->fd); close (pfd->fd); } else { gf_log (this->name, GF_LOG_TRACE, "janitor: closing dir fd=%p", pfd->dir); closedir (pfd->dir); } GF_FREE (pfd); } } return NULL; } void posix_spawn_janitor_thread (xlator_t *this) { struct posix_private *priv = NULL; int ret = 0; priv = this->private; LOCK (&priv->lock); { if (!priv->janitor_present) { ret = pthread_create (&priv->janitor, NULL, posix_janitor_thread_proc, this); if (ret < 0) { gf_log (this->name, GF_LOG_ERROR, "spawning janitor thread failed: %s", strerror (errno)); goto unlock; } priv->janitor_present = _gf_true; } } unlock: UNLOCK (&priv->lock); } static int is_fresh_file (struct stat *stat) { struct timeval tv; gettimeofday (&tv, NULL); if ((stat->st_ctime >= (tv.tv_sec - 1)) && (stat->st_ctime <= tv.tv_sec)) return 1; return 0; } int posix_gfid_heal (xlator_t *this, const char *path, loc_t *loc, dict_t *xattr_req) { /* The purpose of this function is to prevent a race where an inode creation FOP (like mkdir/mknod/create etc) races with lookup in the following way: {create thread} | {lookup thread} | t0 mkdir ("name") | t1 | posix_gfid_set ("name", 2); t2 posix_gfid_set ("name", 1); | t3 lstat ("name"); | lstat ("name"); In the above case mkdir FOP would have resulted with GFID 2 while it should have been GFID 1. It matters in the case where GFID would have gotten set to 1 on other subvolumes of replciate/distribute The "solution" here is that, if we detect lookup is attempting to set a GFID on a file which is created very recently, but does not yet have a GFID (i.e, between t1 and t2), then "fake" it as though posix_gfid_heal was called at t0 instead. */ uuid_t uuid_curr; int ret = 0; struct stat stat = {0, }; if (!xattr_req) goto out; if (sys_lstat (path, &stat) != 0) goto out; ret = sys_lgetxattr (path, GFID_XATTR_KEY, uuid_curr, 16); if (ret != 16) { if (is_fresh_file (&stat)) { ret = -1; errno = ENOENT; goto out; } } ret = posix_gfid_set (this, path, loc, xattr_req); out: return ret; } int posix_acl_xattr_set (xlator_t *this, const char *path, dict_t *xattr_req) { int ret = 0; data_t *data = NULL; struct stat stat = {0, }; if (!xattr_req) goto out; if (sys_lstat (path, &stat) != 0) goto out; data = dict_get (xattr_req, "system.posix_acl_access"); if (data) { ret = sys_lsetxattr (path, "system.posix_acl_access", data->data, data->len, 0); if (ret != 0) goto out; } data = dict_get (xattr_req, "system.posix_acl_default"); if (data) { ret = sys_lsetxattr (path, "system.posix_acl_default", data->data, data->len, 0); if (ret != 0) goto out; } out: return ret; } static int _handle_entry_create_keyvalue_pair (dict_t *d, char *k, data_t *v, void *tmp) { int ret = -1; posix_xattr_filler_t *filler = NULL; filler = tmp; if (!strcmp (GFID_XATTR_KEY, k) || !strcmp ("gfid-req", k) || !strcmp ("system.posix_acl_default", k) || !strcmp ("system.posix_acl_access", k) || ZR_FILE_CONTENT_REQUEST(k)) { return 0; } ret = posix_handle_pair (filler->this, filler->real_path, k, v, XATTR_CREATE); if (ret < 0) { errno = -ret; return -1; } return 0; } int posix_entry_create_xattr_set (xlator_t *this, const char *path, dict_t *dict) { int ret = -1; posix_xattr_filler_t filler = {0,}; if (!dict) goto out; filler.this = this; filler.real_path = path; ret = dict_foreach (dict, _handle_entry_create_keyvalue_pair, &filler); out: return ret; } static int __posix_fd_ctx_get (fd_t *fd, xlator_t *this, struct posix_fd **pfd_p) { uint64_t tmp_pfd = 0; struct posix_fd *pfd = NULL; int ret = -1; char *real_path = NULL; int _fd = -1; DIR *dir = NULL; ret = __fd_ctx_get (fd, this, &tmp_pfd); if (ret == 0) { pfd = (void *)(long) tmp_pfd; ret = 0; goto out; } if (!fd_is_anonymous(fd)) /* anonymous fd */ goto out; MAKE_HANDLE_PATH (real_path, this, fd->inode->gfid, NULL); pfd = GF_CALLOC (1, sizeof (*pfd), gf_posix_mt_posix_fd); if (!pfd) { goto out; } pfd->fd = -1; if (fd->inode->ia_type == IA_IFDIR) { dir = opendir (real_path); if (!dir) { GF_FREE (pfd); pfd = NULL; goto out; } _fd = dirfd (dir); } if (fd->inode->ia_type == IA_IFREG) { _fd = open (real_path, O_RDWR|O_LARGEFILE); if (_fd == -1) { GF_FREE (pfd); pfd = NULL; goto out; } } pfd->fd = _fd; pfd->dir = dir; ret = __fd_ctx_set (fd, this, (uint64_t) (long) pfd); if (ret != 0) { if (_fd != -1) close (_fd); if (dir) closedir (dir); GF_FREE (pfd); pfd = NULL; goto out; } ret = 0; out: if (pfd_p) *pfd_p = pfd; return ret; } int posix_fd_ctx_get (fd_t *fd, xlator_t *this, struct posix_fd **pfd) { int ret; LOCK (&fd->inode->lock); { ret = __posix_fd_ctx_get (fd, this, pfd); } UNLOCK (&fd->inode->lock); return ret; } static void * posix_health_check_thread_proc (void *data) { xlator_t *this = NULL; struct posix_private *priv = NULL; uint32_t interval = 0; int ret = -1; struct stat sb = {0, }; this = data; priv = this->private; /* prevent races when the interval is updated */ interval = priv->health_check_interval; if (interval == 0) goto out; gf_log (this->name, GF_LOG_DEBUG, "health-check thread started, " "interval = %d seconds", interval); while (1) { /* aborting sleep() is a request to exit this thread, sleep() * will normally not return when cancelled */ ret = sleep (interval); if (ret > 0) break; /* prevent thread errors while doing the health-check(s) */ pthread_setcancelstate (PTHREAD_CANCEL_DISABLE, NULL); /* Do the health-check, it should be moved to its own function * in case it gets more complex. */ ret = stat (priv->base_path, &sb); if (ret < 0) { gf_log (this->name, GF_LOG_WARNING, "stat() on %s returned: %s", priv->base_path, strerror (errno)); goto abort; } pthread_setcancelstate (PTHREAD_CANCEL_ENABLE, NULL); } out: gf_log (this->name, GF_LOG_DEBUG, "health-check thread exiting"); LOCK (&priv->lock); { priv->health_check_active = _gf_false; } UNLOCK (&priv->lock); return NULL; abort: /* health-check failed */ gf_log (this->name, GF_LOG_EMERG, "health-check failed, going down"); xlator_notify (this->parents->xlator, GF_EVENT_CHILD_DOWN, this); ret = sleep (30); if (ret == 0) { gf_log (this->name, GF_LOG_EMERG, "still alive! -> SIGTERM"); kill (getpid(), SIGTERM); } ret = sleep (30); if (ret == 0) { gf_log (this->name, GF_LOG_EMERG, "still alive! -> SIGKILL"); kill (getpid(), SIGKILL); } return NULL; } void posix_spawn_health_check_thread (xlator_t *xl) { struct posix_private *priv = NULL; int ret = -1; priv = xl->private; LOCK (&priv->lock); { /* cancel the running thread */ if (priv->health_check_active == _gf_true) { pthread_cancel (priv->health_check); priv->health_check_active = _gf_false; } /* prevent scheduling a check in a tight loop */ if (priv->health_check_interval == 0) goto unlock; ret = pthread_create (&priv->health_check, NULL, posix_health_check_thread_proc, xl); if (ret < 0) { priv->health_check_interval = 0; priv->health_check_active = _gf_false; gf_log (xl->name, GF_LOG_ERROR, "unable to setup health-check thread: %s", strerror (errno)); goto unlock; } /* run the thread detached, resources will be freed on exit */ pthread_detach (priv->health_check); priv->health_check_active = _gf_true; } unlock: UNLOCK (&priv->lock); } int posix_fsyncer_pick (xlator_t *this, struct list_head *head) { struct posix_private *priv = NULL; int count = 0; priv = this->private; pthread_mutex_lock (&priv->fsync_mutex); { while (list_empty (&priv->fsyncs)) pthread_cond_wait (&priv->fsync_cond, &priv->fsync_mutex); count = priv->fsync_queue_count; priv->fsync_queue_count = 0; list_splice_init (&priv->fsyncs, head); } pthread_mutex_unlock (&priv->fsync_mutex); return count; } void posix_fsyncer_process (xlator_t *this, call_stub_t *stub, gf_boolean_t do_fsync) { struct posix_fd *pfd = NULL; int ret = -1; struct posix_private *priv = NULL; priv = this->private; ret = posix_fd_ctx_get (stub->args.fd, this, &pfd); if (ret < 0) { gf_log (this->name, GF_LOG_ERROR, "could not get fdctx for fd(%s)", uuid_utoa (stub->args.fd->inode->gfid)); call_unwind_error (stub, -1, EINVAL); return; } if (do_fsync) { #ifdef HAVE_FDATASYNC if (stub->args.datasync) ret = fdatasync (pfd->fd); else #endif ret = fsync (pfd->fd); } else { ret = 0; } if (ret) { gf_log (this->name, GF_LOG_ERROR, "could not fstat fd(%s)", uuid_utoa (stub->args.fd->inode->gfid)); call_unwind_error (stub, -1, errno); return; } call_unwind_error (stub, 0, 0); } static void posix_fsyncer_syncfs (xlator_t *this, struct list_head *head) { call_stub_t *stub = NULL; struct posix_fd *pfd = NULL; int ret = -1; stub = list_entry (head->prev, call_stub_t, list); ret = posix_fd_ctx_get (stub->args.fd, this, &pfd); if (ret) return; #ifdef GF_LINUX_HOST_OS /* syncfs() is not "declared" in RHEL's glibc even though the kernel has support. */ #include #include #ifdef SYS_syncfs syscall (SYS_syncfs, pfd->fd); #else sync(); #endif #else sync(); #endif } void * posix_fsyncer (void *d) { xlator_t *this = d; struct posix_private *priv = NULL; call_stub_t *stub = NULL; call_stub_t *tmp = NULL; struct list_head list; int count = 0; gf_boolean_t do_fsync = _gf_true; priv = this->private; for (;;) { INIT_LIST_HEAD (&list); count = posix_fsyncer_pick (this, &list); usleep (priv->batch_fsync_delay_usec); gf_log (this->name, GF_LOG_DEBUG, "picked %d fsyncs", count); switch (priv->batch_fsync_mode) { case BATCH_NONE: case BATCH_REVERSE_FSYNC: break; case BATCH_SYNCFS: case BATCH_SYNCFS_SINGLE_FSYNC: case BATCH_SYNCFS_REVERSE_FSYNC: posix_fsyncer_syncfs (this, &list); break; } if (priv->batch_fsync_mode == BATCH_SYNCFS) do_fsync = _gf_false; else do_fsync = _gf_true; list_for_each_entry_safe_reverse (stub, tmp, &list, list) { list_del_init (&stub->list); posix_fsyncer_process (this, stub, do_fsync); if (priv->batch_fsync_mode == BATCH_SYNCFS_SINGLE_FSYNC) do_fsync = _gf_false; } } }