From 554f7b8a773b3cffd74707c18172366ab45ea306 Mon Sep 17 00:00:00 2001 From: Neels Hofmeyr Date: Wed, 20 Dec 2017 01:14:31 +0100 Subject: [PATCH] rate_ctr: fix osmo-sgsn DoS: don't return NULL on already used index Recent patch I563764af1d28043e909234ebb048239125ce6ecd introduced returning NULL from rate_ctr_group_alloc() when the index passed already exists. Instead of returning NULL, find an unused group index and use that, adjust the error message. In stats_test.c, adjust, and also assert allocated counter group indexes everywhere. Rationale: The original patch causes osmo-sgsn to crash as soon as the second subscriber attempts to establish an MM context. Of course osmo-sgsn is wrong to a) fail to check a NULL return value and crash and b) to fail to allocate an MM context just because the rate counter group could not be allocated (it still rejects the MM context completely if rate_ctr_group_alloc() fails). Nevertheless, the price we pay for rate counter correctness is, at least in this instance, way too high: osmo-sgsn becomes completely unusable for more than one subscriber. Numerous other places exist where rate_ctr_group_alloc() is called with a constant index number; from a quick grep magic I found these possible breaking points: osmo-sgsn/src/gprs/gb_proxy.c:1431: cfg->ctrg = rate_ctr_group_alloc(tall_bsc_ctx, &global_ctrg_desc, 0); osmo-sgsn/src/gprs/gprs_sgsn.c:139: sgsn->rate_ctrs = rate_ctr_group_alloc(tall_bsc_ctx, &sgsn_ctrg_desc, 0); osmo-sgsn/src/gprs/gprs_sgsn.c:270: ctx->ctrg = rate_ctr_group_alloc(ctx, &mmctx_ctrg_desc, 0); osmo-sgsn/src/gprs/gtphub.c:888: b->counters_io = rate_ctr_group_alloc(osmo_gtphub_ctx, >phub_ctrg_io_desc, 0); osmo-bsc/src/libfilter/bsc_msg_acc.c:87: lst->stats = rate_ctr_group_alloc(lst, &bsc_cfg_acc_list_desc, 0); osmo-pcu/src/bts.cpp:228: m_ratectrs = rate_ctr_group_alloc(tall_pcu_ctx, &bts_ctrg_desc, 0); osmo-pcu/src/tbf.cpp:793: tbf->m_ctrs = rate_ctr_group_alloc(tbf, &tbf_ctrg_desc, 0); osmo-pcu/src/tbf.cpp:879: tbf->m_ul_egprs_ctrs = rate_ctr_group_alloc(tbf, &tbf_ul_egprs_ctrg_desc, 0); osmo-pcu/src/tbf.cpp:880: tbf->m_ul_gprs_ctrs = rate_ctr_group_alloc(tbf, &tbf_ul_gprs_ctrg_desc, 0); osmo-pcu/src/tbf.cpp:970: tbf->m_dl_egprs_ctrs = rate_ctr_group_alloc(tbf, &tbf_dl_egprs_ctrg_desc, 0); osmo-pcu/src/tbf.cpp:977: tbf->m_dl_gprs_ctrs = rate_ctr_group_alloc(tbf, &tbf_dl_gprs_ctrg_desc, 0); osmo-pcu/src/tbf.cpp:1475: ul_tbf->m_ctrs = rate_ctr_group_alloc(ul_tbf, &tbf_ctrg_desc, 0); osmo-pcu/src/bts.cpp:226: m_ratectrs = rate_ctr_group_alloc(tall_pcu_ctx, &bts_ctrg_desc, 1); We can fix all of these callers and then reconsider returning NULL, but IMO even into the future, rate counter group indexes are not something worth failing to provide service for. For future bugs we should keep the automatic index picking in case of index collisions. We will get an error message barfed and can fix the issue in our own time, while the application remains completely usable, and even the rate counters can still be queried (at wrong indexes, but life is tough). Related: I49aa95b610f2faec52dede2e4816da47ca1dfb14 (osmo-sgsn's segfault) Change-Id: Iba6e41b8eeaea5ff6ed862bab3f34a62ab976914 --- src/rate_ctr.c | 29 ++++++++++++++++++++++++++--- tests/stats/stats_test.c | 10 +++++----- 2 files changed, 31 insertions(+), 8 deletions(-) diff --git a/src/rate_ctr.c b/src/rate_ctr.c index 72c6bb118..ec054c575 100644 --- a/src/rate_ctr.c +++ b/src/rate_ctr.c @@ -178,6 +178,27 @@ err_free: return NULL; } +/*! Find an unused index for this rate counter group. + * \param[in] name Name of the counter group + * \returns the largest used index number + 1, or 0 if none exist yet. */ +static unsigned int rate_ctr_get_unused_name_idx(const char *name) +{ + unsigned int idx = 0; + struct rate_ctr_group *ctrg; + + llist_for_each_entry(ctrg, &rate_ctr_groups, list) { + if (!ctrg->desc) + continue; + + if (strcmp(ctrg->desc->group_name_prefix, name)) + continue; + + if (idx <= ctrg->idx) + idx = ctrg->idx + 1; + } + return idx; +} + /*! Allocate a new group of counters according to description * \param[in] ctx \ref talloc context * \param[in] desc Rate counter group description @@ -191,9 +212,11 @@ struct rate_ctr_group *rate_ctr_group_alloc(void *ctx, struct rate_ctr_group *group; if (rate_ctr_get_group_by_name_idx(desc->group_name_prefix, idx)) { - LOGP(DLGLOBAL, LOGL_ERROR, "counter group '%s' already exists for index %u\n", - desc->group_name_prefix, idx); - return NULL; /* group already exist */ + unsigned int new_idx = rate_ctr_get_unused_name_idx(desc->group_name_prefix); + LOGP(DLGLOBAL, LOGL_ERROR, "counter group '%s' already exists for index %u," + " instead using index %u. This is a software bug that needs fixing.\n", + desc->group_name_prefix, idx, new_idx); + idx = new_idx; } size = sizeof(struct rate_ctr_group) + diff --git a/tests/stats/stats_test.c b/tests/stats/stats_test.c index 35faf9a6d..6ef884181 100644 --- a/tests/stats/stats_test.c +++ b/tests/stats/stats_test.c @@ -324,16 +324,16 @@ static void test_reporting() statg2 = osmo_stat_item_group_alloc(stats_ctx, &statg_desc, 2); OSMO_ASSERT(statg2 != NULL); ctrg1 = rate_ctr_group_alloc(stats_ctx, &ctrg_desc, 1); - OSMO_ASSERT(ctrg1 != NULL); + OSMO_ASSERT(ctrg1 && ctrg1->idx == 1); ctrg2 = rate_ctr_group_alloc(stats_ctx, &ctrg_desc, 2); - OSMO_ASSERT(ctrg2 != NULL); + OSMO_ASSERT(ctrg2 && ctrg2->idx == 2); ctrg_dup = rate_ctr_group_alloc(stats_ctx, &ctrg_desc, 2); - if (ctrg_dup != NULL && ctrg2 != NULL) - printf("FAIL: successfully allocated already existing counter group!\n"); + OSMO_ASSERT(ctrg_dup && ctrg_dup->idx == 3); + rate_ctr_group_free(ctrg_dup); ctrg3 = rate_ctr_group_alloc(stats_ctx, &ctrg_desc_dot, 3); - OSMO_ASSERT(ctrg3 != NULL); + OSMO_ASSERT(ctrg3 && ctrg3->idx == 3); srep1 = stats_reporter_create_test("test1"); OSMO_ASSERT(srep1 != NULL);