From: Mark Stapp <mjs@voltanet.io>
Date: Thu, 13 Dec 2018 19:15:27 +0000 (-0500)
Subject: zebra: use a small retry timeout for the rib workqueue
X-Git-Tag: frr-7.1-dev~93^2
X-Git-Url: https://git.puffer.fish/?a=commitdiff_plain;h=refs%2Fpull%2F3464%2Fhead;p=mirror%2Ffrr.git

zebra: use a small retry timeout for the rib workqueue

In the zebra rib processing workqueue, set a small timeout
so that we will wait a short time if the queue into the
async dataplane is full. This helps avoid a situation where
the zebra main pthread constantly retries rib work without
giving the dataplane pthread a chance to make progress.

Signed-off-by: Mark Stapp <mjs@voltanet.io>
---

diff --git a/zebra/zebra_rib.c b/zebra/zebra_rib.c
index f2d07310ee..9f1374af57 100644
--- a/zebra/zebra_rib.c
+++ b/zebra/zebra_rib.c
@@ -2362,6 +2362,7 @@ static void rib_queue_init(struct zebra_t *zebra)
 	/* XXX: TODO: These should be runtime configurable via vty */
 	zebra->ribq->spec.max_retries = 3;
 	zebra->ribq->spec.hold = ZEBRA_RIB_PROCESS_HOLD_TIME;
+	zebra->ribq->spec.retry = ZEBRA_RIB_PROCESS_RETRY_TIME;
 
 	if (!(zebra->mq = meta_queue_new())) {
 		flog_err(EC_ZEBRA_WQ_NONEXISTENT,
diff --git a/zebra/zserv.h b/zebra/zserv.h
index f7967f54f0..fe1dbdbfe4 100644
--- a/zebra/zserv.h
+++ b/zebra/zserv.h
@@ -184,6 +184,7 @@ struct zebra_t {
 
 /* rib work queue */
 #define ZEBRA_RIB_PROCESS_HOLD_TIME 10
+#define ZEBRA_RIB_PROCESS_RETRY_TIME 5
 	struct work_queue *ribq;
 	struct meta_queue *mq;