[GCM] Update backoff policy to be more conservative.

[GCM] Update backoff policy to be more conservative.

Just tweaking the backoff constants to better reflect what to do in case of
server error. Future patches will introduce alternative policies for different
kinds of errors.

BUG=284553
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=251462

[Nicolas Zea, 2014-02-14 18:33:03]

PTAL

[fgorski, 2014-02-14 18:36:28]

lgtm,

please take a look at the comments

https://codereview.chromium.org/164183008/diff/1/google_apis/gcm/gcm_client_i...
File google_apis/gcm/gcm_client_impl.cc (right):

https://codereview.chromium.org/164183008/diff/1/google_apis/gcm/gcm_client_i...
google_apis/gcm/gcm_client_impl.cc:30: // Backoff policy. Shared across
reconnection logic and checkin/registration
also unregistration

https://codereview.chromium.org/164183008/diff/1/google_apis/gcm/gcm_client_i...
google_apis/gcm/gcm_client_impl.cc:33: // server reasons), we have a 30s += 10s
(33%) jitter initial backoff.
+/- insted of +=?

https://codereview.chromium.org/164183008/diff/1/google_apis/gcm/gcm_client_i...
google_apis/gcm/gcm_client_impl.cc:42: 1000 * 30,  // 30 seconds.
nit: would big-endian be more intuitive?
30* 1000

https://codereview.chromium.org/164183008/diff/1/google_apis/gcm/gcm_client_i...
google_apis/gcm/gcm_client_impl.cc:52: 1000 * 60 * 10, // 10 minutes.
nit: same here, big endian?

[jianli, 2014-02-14 18:38:07]

lgtm

[Nicolas Zea, 2014-02-14 18:39:46]

Done.

https://codereview.chromium.org/164183008/diff/1/google_apis/gcm/gcm_client_i...
File google_apis/gcm/gcm_client_impl.cc (right):

https://codereview.chromium.org/164183008/diff/1/google_apis/gcm/gcm_client_i...
google_apis/gcm/gcm_client_impl.cc:30: // Backoff policy. Shared across
reconnection logic and checkin/registration
On 2014/02/14 18:36:29, Filip Gorski wrote:
> also unregistration

Done.

https://codereview.chromium.org/164183008/diff/1/google_apis/gcm/gcm_client_i...
google_apis/gcm/gcm_client_impl.cc:33: // server reasons), we have a 30s += 10s
(33%) jitter initial backoff.
On 2014/02/14 18:36:29, Filip Gorski wrote:
> +/- insted of +=?

Done.

https://codereview.chromium.org/164183008/diff/1/google_apis/gcm/gcm_client_i...
google_apis/gcm/gcm_client_impl.cc:42: 1000 * 30,  // 30 seconds.
On 2014/02/14 18:36:29, Filip Gorski wrote:
> nit: would big-endian be more intuitive?
> 30* 1000

Done.

https://codereview.chromium.org/164183008/diff/1/google_apis/gcm/gcm_client_i...
google_apis/gcm/gcm_client_impl.cc:52: 1000 * 60 * 10, // 10 minutes.
On 2014/02/14 18:36:29, Filip Gorski wrote:
> nit: same here, big endian?

Done.

[Nicolas Zea, 2014-02-14 18:39:50]

The CQ bit was checked by zea@chromium.org

[commit-bot: I haz the power, 2014-02-14 18:41:35]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/zea@chromium.org/164183008/70001

[commit-bot: I haz the power, 2014-02-14 20:22:18]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-02-14 20:22:18]

Retried try job too often on win_rel for step(s) unit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

[Nicolas Zea, 2014-02-14 20:45:44]

The CQ bit was checked by zea@chromium.org

[commit-bot: I haz the power, 2014-02-14 20:47:20]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/zea@chromium.org/164183008/70001

[commit-bot: I haz the power, 2014-02-15 00:55:01]

Change committed as 251462